mirror of
https://github.com/anomalyco/opencode.git
synced 2026-08-15 17:08:21 -04:00
Compare commits
3 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 988efb6217 | |||
| e1d1352d8f | |||
| ab5bd520e2 |
@@ -1,5 +0,0 @@
|
||||
---
|
||||
"@opencode-ai/core": patch
|
||||
---
|
||||
|
||||
Preserve prompt cache prefixes when sessions move between locations with unchanged instructions.
|
||||
@@ -1,5 +0,0 @@
|
||||
---
|
||||
"@opencode-ai/plugin": patch
|
||||
---
|
||||
|
||||
Derive Promise plugin API request and response conversion from the canonical protocol schemas.
|
||||
@@ -1,8 +0,0 @@
|
||||
---
|
||||
"@opencode-ai/core": minor
|
||||
"@opencode-ai/schema": minor
|
||||
"@opencode-ai/protocol": minor
|
||||
"@opencode-ai/client": minor
|
||||
---
|
||||
|
||||
Remove the unused question request API and use session forms for question tool interactions.
|
||||
@@ -1,9 +0,0 @@
|
||||
---
|
||||
"@opencode-ai/core": minor
|
||||
"@opencode-ai/schema": minor
|
||||
"@opencode-ai/protocol": minor
|
||||
"@opencode-ai/client": minor
|
||||
---
|
||||
|
||||
Add an opt-in portable shell permission scanner. Opaque commands use normal shell authorization without inferring
|
||||
external directories, while the default tree-sitter path remains unchanged.
|
||||
@@ -1,4 +1,2 @@
|
||||
packages/core/migration/**/snapshot.json linguist-generated
|
||||
packages/core/src/database/migration.gen.ts linguist-generated
|
||||
packages/core/src/models-dev/snapshot.txt linguist-generated
|
||||
packages/core/src/**/*.txt text eol=lf
|
||||
|
||||
@@ -1,5 +1,4 @@
|
||||
adamdotdevin
|
||||
arvsrn
|
||||
Brendonovich
|
||||
fwang
|
||||
Hona
|
||||
@@ -8,15 +7,11 @@ jayair
|
||||
jlongster
|
||||
kitlangton
|
||||
kommander
|
||||
ludvigrask
|
||||
MrMushrooooom
|
||||
neriousy
|
||||
nexxeln
|
||||
R44VC0RP
|
||||
rekram1-node
|
||||
thdxr
|
||||
simonklee
|
||||
Slickstef11
|
||||
usrnk1
|
||||
vimtor
|
||||
StarpTech
|
||||
starptech
|
||||
|
||||
@@ -1,37 +0,0 @@
|
||||
name: deploy-www
|
||||
|
||||
on:
|
||||
push:
|
||||
branches:
|
||||
- dev
|
||||
- v2
|
||||
workflow_dispatch:
|
||||
|
||||
concurrency:
|
||||
group: deploy-www-${{ github.ref_name }}
|
||||
cancel-in-progress: false
|
||||
|
||||
permissions:
|
||||
contents: read
|
||||
|
||||
jobs:
|
||||
deploy:
|
||||
if: github.repository == 'anomalyco/opencode' && (github.ref_name == 'dev' || github.ref_name == 'v2')
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- uses: actions/checkout@f43a0e5ff2bd294095638e18286ca9a3d1956744 # v3.6.0
|
||||
|
||||
- uses: ./.github/actions/setup-bun
|
||||
|
||||
- name: Build
|
||||
working-directory: packages/www
|
||||
run: bun run build
|
||||
env:
|
||||
BLUME_ENV: ${{ github.ref_name == 'v2' && 'production' || 'dev' }}
|
||||
CLOUDFLARE_ENV: ${{ github.ref_name == 'v2' && 'production' || 'dev' }}
|
||||
|
||||
- name: Deploy
|
||||
working-directory: packages/www
|
||||
run: bun run deploy
|
||||
env:
|
||||
CLOUDFLARE_API_TOKEN: ${{ secrets.CLOUDFLARE_API_TOKEN }}
|
||||
@@ -4,7 +4,6 @@ on:
|
||||
push:
|
||||
branches:
|
||||
- dev
|
||||
- v2
|
||||
|
||||
jobs:
|
||||
generate:
|
||||
|
||||
+16
-136
@@ -6,6 +6,7 @@ on:
|
||||
branches:
|
||||
- ci
|
||||
- dev
|
||||
- v2
|
||||
- beta
|
||||
- fix/npm-native-binary-install
|
||||
- snapshot-*
|
||||
@@ -74,7 +75,7 @@ jobs:
|
||||
build-cli:
|
||||
needs: version
|
||||
runs-on: blacksmith-4vcpu-ubuntu-2404
|
||||
if: github.repository == 'anomalyco/opencode' && github.ref_name != 'beta'
|
||||
if: github.repository == 'anomalyco/opencode'
|
||||
steps:
|
||||
- uses: actions/checkout@f43a0e5ff2bd294095638e18286ca9a3d1956744 # v3.6.0
|
||||
with:
|
||||
@@ -89,18 +90,11 @@ jobs:
|
||||
opencode-app-id: ${{ vars.OPENCODE_APP_ID }}
|
||||
opencode-app-secret: ${{ secrets.OPENCODE_APP_SECRET }}
|
||||
|
||||
- name: Build legacy CLI
|
||||
if: github.ref_name != 'v2' && github.ref_name != 'beta'
|
||||
run: ./packages/opencode/script/build.ts ${{ (github.ref_name == 'beta' && '--sourcemaps') || '' }}
|
||||
env:
|
||||
OPENCODE_VERSION: ${{ needs.version.outputs.version }}
|
||||
OPENCODE_RELEASE: ${{ needs.version.outputs.release }}
|
||||
GH_REPO: ${{ needs.version.outputs.repo }}
|
||||
GH_TOKEN: ${{ steps.committer.outputs.token }}
|
||||
|
||||
- name: Build preview CLI
|
||||
- name: Build
|
||||
id: build
|
||||
run: ./packages/cli/script/build.ts ${{ (github.ref_name == 'beta' && '--sourcemaps') || '' }}
|
||||
run: |
|
||||
./packages/opencode/script/build.ts ${{ (github.ref_name == 'beta' && '--sourcemaps') || '' }}
|
||||
./packages/cli/script/build.ts ${{ (github.ref_name == 'beta' && '--sourcemaps') || '' }}
|
||||
env:
|
||||
OPENCODE_VERSION: ${{ needs.version.outputs.version }}
|
||||
OPENCODE_RELEASE: ${{ needs.version.outputs.release }}
|
||||
@@ -108,7 +102,6 @@ jobs:
|
||||
GH_TOKEN: ${{ steps.committer.outputs.token }}
|
||||
|
||||
- uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
|
||||
if: github.ref_name != 'v2' && github.ref_name != 'beta'
|
||||
with:
|
||||
name: opencode-cli
|
||||
path: |
|
||||
@@ -116,128 +109,24 @@ jobs:
|
||||
packages/opencode/dist/opencode-linux*
|
||||
|
||||
- uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
|
||||
if: github.ref_name != 'v2' && github.ref_name != 'beta'
|
||||
with:
|
||||
name: opencode-cli-windows
|
||||
path: packages/opencode/dist/opencode-windows*
|
||||
|
||||
- uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
|
||||
with:
|
||||
name: opencode-preview-cli-unsigned
|
||||
name: opencode-preview-cli
|
||||
path: packages/cli/dist/cli-*
|
||||
|
||||
outputs:
|
||||
version: ${{ needs.version.outputs.version }}
|
||||
|
||||
sign-cli-macos:
|
||||
needs: build-cli
|
||||
runs-on: macos-26
|
||||
if: github.repository == 'anomalyco/opencode'
|
||||
steps:
|
||||
- uses: actions/checkout@f43a0e5ff2bd294095638e18286ca9a3d1956744 # v3.6.0
|
||||
|
||||
- uses: apple-actions/import-codesign-certs@8f3fb608891dd2244cdab3d69cd68c0d37a7fe93 # v2.0.0
|
||||
with:
|
||||
keychain: build
|
||||
p12-file-base64: ${{ secrets.APPLE_CERTIFICATE }}
|
||||
p12-password: ${{ secrets.APPLE_CERTIFICATE_PASSWORD }}
|
||||
|
||||
- uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
|
||||
with:
|
||||
name: opencode-preview-cli-unsigned
|
||||
path: packages/cli/dist
|
||||
|
||||
- name: Sign macOS CLI binaries
|
||||
run: |
|
||||
identity=$(security find-identity -v -p codesigning build.keychain | sed -n 's/.*"\(Developer ID Application:.*\)"/\1/p' | head -n 1)
|
||||
if [ -z "$identity" ]; then
|
||||
echo "Developer ID Application identity not found"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
found=0
|
||||
for file in packages/cli/dist/cli-darwin-*/bin/opencode2; do
|
||||
if [ ! -f "$file" ]; then
|
||||
continue
|
||||
fi
|
||||
found=1
|
||||
codesign \
|
||||
--force \
|
||||
--timestamp \
|
||||
--options runtime \
|
||||
--entitlements packages/cli/script/entitlements.plist \
|
||||
--sign "$identity" \
|
||||
"$file"
|
||||
codesign --verify --deep --strict --verbose=4 "$file"
|
||||
codesign --display --requirements - "$file"
|
||||
done
|
||||
|
||||
if [ "$found" -eq 0 ]; then
|
||||
echo "No macOS CLI binaries found"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
- uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
|
||||
with:
|
||||
name: opencode-preview-cli
|
||||
path: packages/cli/dist/cli-*
|
||||
if-no-files-found: error
|
||||
|
||||
build-node-cli:
|
||||
needs: version
|
||||
if: github.repository == 'anomalyco/opencode' && github.ref_name != 'beta'
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
settings:
|
||||
- target: linux-arm64
|
||||
host: blacksmith-4vcpu-ubuntu-2404-arm
|
||||
- target: linux-x64
|
||||
host: blacksmith-4vcpu-ubuntu-2404
|
||||
- target: darwin-arm64
|
||||
host: macos-26
|
||||
- target: windows-arm64
|
||||
host: blacksmith-4vcpu-windows-2025
|
||||
- target: windows-x64
|
||||
host: blacksmith-4vcpu-windows-2025
|
||||
runs-on: ${{ matrix.settings.host }}
|
||||
defaults:
|
||||
run:
|
||||
shell: bash
|
||||
steps:
|
||||
- uses: actions/checkout@f43a0e5ff2bd294095638e18286ca9a3d1956744 # v3.6.0
|
||||
|
||||
- uses: ./.github/actions/setup-bun
|
||||
with:
|
||||
install-flags: --os=* --cpu=*
|
||||
|
||||
- uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020 # v4.4.0
|
||||
with:
|
||||
node-version: "26.4.0"
|
||||
|
||||
- name: Build
|
||||
run: bun packages/cli/script/build-node.ts --target=${{ matrix.settings.target }} --skip-install --outdir=dist/node
|
||||
env:
|
||||
OPENCODE_VERSION: ${{ needs.version.outputs.version }}
|
||||
OPENCODE_RELEASE: ${{ needs.version.outputs.release }}
|
||||
|
||||
- name: Verify service lifecycle
|
||||
if: matrix.settings.target != 'windows-arm64'
|
||||
working-directory: packages/cli
|
||||
run: bun run script/service-smoke.ts --node
|
||||
|
||||
- uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
|
||||
with:
|
||||
name: opencode-node-cli-${{ matrix.settings.target }}
|
||||
path: packages/cli/dist/node/cli-node-*
|
||||
if-no-files-found: error
|
||||
|
||||
sign-cli-windows:
|
||||
needs:
|
||||
- build-cli
|
||||
- version
|
||||
runs-on: blacksmith-4vcpu-windows-2025
|
||||
if: github.repository == 'anomalyco/opencode' && github.ref_name != 'v2' && github.ref_name != 'beta'
|
||||
if: github.repository == 'anomalyco/opencode' && github.ref_name != 'v2'
|
||||
env:
|
||||
AZURE_CLIENT_ID: ${{ secrets.AZURE_CLIENT_ID }}
|
||||
AZURE_TENANT_ID: ${{ secrets.AZURE_TENANT_ID }}
|
||||
@@ -334,6 +223,7 @@ jobs:
|
||||
|
||||
build-electron:
|
||||
needs:
|
||||
- build-cli
|
||||
- version
|
||||
if: github.repository == 'anomalyco/opencode' && github.ref_name != 'v2'
|
||||
continue-on-error: false
|
||||
@@ -430,14 +320,16 @@ jobs:
|
||||
env:
|
||||
OPENCODE_VERSION: ${{ needs.version.outputs.version }}
|
||||
OPENCODE_CHANNEL: ${{ (github.ref_name == 'beta' && 'beta') || 'prod' }}
|
||||
OPENCODE_CLI_TARGET: ${{ matrix.settings.target }}
|
||||
OPENCODE_CLI_ARTIFACT: ${{ (runner.os == 'Windows' && 'opencode-cli-windows') || 'opencode-cli' }}
|
||||
RUST_TARGET: ${{ matrix.settings.target }}
|
||||
GH_TOKEN: ${{ github.token }}
|
||||
GITHUB_RUN_ID: ${{ github.run_id }}
|
||||
|
||||
- name: Build
|
||||
run: bun run build
|
||||
working-directory: packages/desktop
|
||||
env:
|
||||
NODE_OPTIONS: --max-old-space-size=4096
|
||||
OPENCODE_VERSION: ${{ needs.version.outputs.version }}
|
||||
OPENCODE_CHANNEL: ${{ (github.ref_name == 'beta' && 'beta') || 'prod' }}
|
||||
SENTRY_AUTH_TOKEN: ${{ secrets.SENTRY_AUTH_TOKEN }}
|
||||
SENTRY_ORG: ${{ vars.SENTRY_ORG }}
|
||||
@@ -446,7 +338,6 @@ jobs:
|
||||
VITE_SENTRY_DSN: ${{ vars.WEB_SENTRY_DSN }}
|
||||
VITE_SENTRY_ENVIRONMENT: ${{ (github.ref_name == 'beta' && 'beta') || 'production' }}
|
||||
VITE_SENTRY_RELEASE: desktop@${{ needs.version.outputs.version }}
|
||||
OPENCODE_CLI_TARGET: ${{ matrix.settings.target }}
|
||||
|
||||
- name: Package
|
||||
if: needs.version.outputs.release
|
||||
@@ -456,7 +347,8 @@ jobs:
|
||||
env:
|
||||
OPENCODE_CHANNEL: ${{ (github.ref_name == 'beta' && 'beta') || 'prod' }}
|
||||
GH_TOKEN: ${{ steps.committer.outputs.token }}
|
||||
CSC_KEYCHAIN: build.keychain
|
||||
CSC_LINK: ${{ secrets.APPLE_CERTIFICATE }}
|
||||
CSC_KEY_PASSWORD: ${{ secrets.APPLE_CERTIFICATE_PASSWORD }}
|
||||
APPLE_API_KEY: ${{ runner.temp }}/apple-api-key.p8
|
||||
APPLE_API_KEY_ID: ${{ secrets.APPLE_API_KEY }}
|
||||
APPLE_API_ISSUER: ${{ secrets.APPLE_API_ISSUER }}
|
||||
@@ -521,8 +413,6 @@ jobs:
|
||||
needs:
|
||||
- version
|
||||
- build-cli
|
||||
- sign-cli-macos
|
||||
- build-node-cli
|
||||
- sign-cli-windows
|
||||
- build-electron
|
||||
if: always() && !failure() && !cancelled()
|
||||
@@ -551,36 +441,26 @@ jobs:
|
||||
registry-url: "https://registry.npmjs.org"
|
||||
|
||||
- uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
|
||||
if: github.ref_name != 'v2' && github.ref_name != 'beta'
|
||||
with:
|
||||
name: opencode-cli
|
||||
path: packages/opencode/dist
|
||||
|
||||
- uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
|
||||
if: github.ref_name != 'v2' && github.ref_name != 'beta'
|
||||
with:
|
||||
name: opencode-cli-windows
|
||||
path: packages/opencode/dist
|
||||
|
||||
- uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
|
||||
if: github.ref_name != 'v2' && github.ref_name != 'beta'
|
||||
if: github.ref_name != 'v2'
|
||||
with:
|
||||
name: opencode-cli-signed-windows
|
||||
path: packages/opencode/dist
|
||||
|
||||
- uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
|
||||
if: github.ref_name != 'beta'
|
||||
with:
|
||||
name: opencode-preview-cli
|
||||
path: packages/cli/dist
|
||||
|
||||
- uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
|
||||
if: github.ref_name != 'beta'
|
||||
with:
|
||||
pattern: opencode-node-cli-*
|
||||
path: packages/cli/dist/node
|
||||
merge-multiple: true
|
||||
|
||||
- uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
|
||||
if: needs.version.outputs.release
|
||||
with:
|
||||
|
||||
@@ -55,12 +55,6 @@ jobs:
|
||||
git config --global user.email "bot@opencode.ai"
|
||||
git config --global user.name "opencode"
|
||||
|
||||
- name: Install ffmpeg
|
||||
if: runner.os == 'Linux'
|
||||
run: |
|
||||
sudo apt-get update
|
||||
sudo apt-get install --yes ffmpeg
|
||||
|
||||
- name: Cache Turbo
|
||||
uses: actions/cache@0057852bfaa89a56745cba8c7296529d2fc39830 # v4.3.0
|
||||
with:
|
||||
@@ -76,38 +70,11 @@ jobs:
|
||||
env:
|
||||
OPENCODE_EXPERIMENTAL_DISABLE_FILEWATCHER: ${{ runner.os == 'Windows' && 'true' || 'false' }}
|
||||
|
||||
- name: Verify compiled service lifecycle
|
||||
if: always()
|
||||
timeout-minutes: 10
|
||||
working-directory: packages/cli
|
||||
run: |
|
||||
bun run script/build.ts --single --skip-install
|
||||
bun run script/service-smoke.ts
|
||||
|
||||
- name: Setup Node build runtime
|
||||
if: always()
|
||||
uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020 # v4.4.0
|
||||
with:
|
||||
node-version: "26.4.0"
|
||||
|
||||
- name: Verify Node build
|
||||
if: always()
|
||||
timeout-minutes: 15
|
||||
working-directory: packages/cli
|
||||
run: |
|
||||
bun run script/build-node.ts --single --skip-install --outdir=dist/node
|
||||
bun run script/service-smoke.ts --node
|
||||
|
||||
- name: Check generated client
|
||||
if: runner.os == 'Linux'
|
||||
working-directory: packages/client
|
||||
run: bun run check:generated
|
||||
|
||||
- name: Check generated documentation
|
||||
if: runner.os == 'Linux'
|
||||
working-directory: packages/www
|
||||
run: bun run check:generated
|
||||
|
||||
e2e:
|
||||
name: e2e (${{ matrix.settings.name }})
|
||||
if: github.ref_name != 'v2' && github.head_ref != 'v2'
|
||||
|
||||
@@ -11,7 +11,6 @@ node_modules
|
||||
playground
|
||||
tmp
|
||||
dist
|
||||
dist-node
|
||||
ts-dist
|
||||
.turbo
|
||||
.typecheck-profiles
|
||||
@@ -26,7 +25,6 @@ Session.vim
|
||||
a.out
|
||||
target
|
||||
.scripts
|
||||
.cache
|
||||
.direnv/
|
||||
|
||||
# Local dev files
|
||||
|
||||
@@ -2,7 +2,7 @@
|
||||
description: "Bump AI sdk dependencies minor / patch versions only"
|
||||
---
|
||||
|
||||
Please read @package.json and @packages/core/package.json.
|
||||
Please read @package.json and @packages/opencode/package.json.
|
||||
|
||||
Your job is to look into AI SDK dependencies, figure out if they have versions that can be upgraded (minor or patch versions ONLY no major ignore major changes).
|
||||
|
||||
|
||||
@@ -6,7 +6,15 @@ subtask: true
|
||||
|
||||
commit and push
|
||||
|
||||
Use `type(scope): summary` with one of these types: `feat`, `fix`, `docs`, `chore`, `refactor`, or `test`. The scope is optional.
|
||||
make sure it includes a prefix like
|
||||
docs:
|
||||
tui:
|
||||
core:
|
||||
ci:
|
||||
ignore:
|
||||
wip:
|
||||
|
||||
For anything in the packages/web use the docs: prefix.
|
||||
|
||||
prefer to explain WHY something was done from an end user perspective instead of
|
||||
WHAT was done.
|
||||
|
||||
@@ -2,7 +2,7 @@
|
||||
description: Remove AI code slop
|
||||
---
|
||||
|
||||
Check the diff against `origin/v2`, and remove all AI generated slop introduced in this branch.
|
||||
Check the diff against dev, and remove all AI generated slop introduced in this branch.
|
||||
|
||||
This includes:
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
---
|
||||
description: translate English to other languages
|
||||
model: opencode/gpt-5.6-sol
|
||||
model: opencode/claude-opus-4-8
|
||||
---
|
||||
|
||||
run git diff and translate changed english doc and UI copy files to other international languages. Translate all languages in parallel to save time.
|
||||
|
||||
@@ -1,48 +0,0 @@
|
||||
# he Glossary
|
||||
|
||||
## Sources
|
||||
|
||||
- Hebrew Academy approved IT terminology: https://terms.hebrew-academy.org.il/Millonim/ShowMillon?KodMillon=192
|
||||
- Firefox Hebrew localization corpus: https://github.com/mozilla-l10n/firefox-l10n/tree/main/he
|
||||
- KDE Hebrew localization team and corpus: https://l10n.kde.org/team-infos.php?teamcode=he
|
||||
- Community-maintained VS Code Hebrew language pack: https://github.com/AMAARETS/vscode-language-pack-he
|
||||
- Microsoft Hebrew developer documentation for Git terminology: https://learn.microsoft.com/he-il/power-platform/alm/tutorials/github-actions-deploy
|
||||
- W3C guidance for bidirectional text: https://www.w3.org/International/articles/strings-and-bidi/
|
||||
|
||||
## Do Not Translate (Locale Additions)
|
||||
|
||||
- `OpenCode` (preserve casing in prose and UI copy)
|
||||
- `API`, `MCP`, `LSP`, `OAuth`, `Git`, model names, and provider names
|
||||
- Commands, flags, keyboard shortcuts, file paths, URLs, identifiers, hashes, and code literals
|
||||
- Keep `commit` and `diff` when they name the exact Git artifact or operation
|
||||
|
||||
## Preferred Terms
|
||||
|
||||
| English / Context | Preferred | Notes |
|
||||
| ----------------- | ------------- | ------------------------------------------------------------------------- |
|
||||
| session | `הפעלה` | Use `שיחה` only when the source specifically means a chat or conversation |
|
||||
| workspace | `סביבת עבודה` | |
|
||||
| terminal | `מסוף` | Prefer the established Hebrew term over transliteration |
|
||||
| command | `פקודה` | |
|
||||
| provider | `ספק` | Use `ספק מודלים` where the bare noun is ambiguous |
|
||||
| model | `מודל` | |
|
||||
| API key | `מפתח API` | Keep the acronym in Latin letters |
|
||||
| plugin | `תוסף` | |
|
||||
| repository | `מאגר` | Use `מאגר Git` where context is ambiguous |
|
||||
| branch | `ענף` | |
|
||||
| context | `הקשר` | Use `חלון הקשר` for context window |
|
||||
| tokens | `אסימונים` | |
|
||||
|
||||
## Guidance
|
||||
|
||||
- Prefer natural modern Israeli Hebrew over word-for-word translation or obscure coined terms.
|
||||
- Use short action verbs for controls and translate complete phrases in context.
|
||||
- Keep recognized developer acronyms and exact Git vocabulary in Latin script instead of phonetic transliteration.
|
||||
- Treat embedded code, paths, commands, shortcuts, hashes, model IDs, and other Latin technical artifacts as LTR content inside the RTL interface.
|
||||
- Keep recurring concepts consistent and do not collapse session, chat, run, and launch into one Hebrew term.
|
||||
|
||||
## Avoid
|
||||
|
||||
- Avoid transliterations such as `טרמינל`, `פלאגין`, and `קומנד` when `מסוף`, `תוסף`, and `פקודה` are clear.
|
||||
- Avoid translating `commit` as `התחייבות`.
|
||||
- Avoid inventing Hebrew expansions for `API`, `MCP`, or `LSP`.
|
||||
@@ -1,6 +1,6 @@
|
||||
---
|
||||
name: effect
|
||||
description: Work with Effect v4 TypeScript code in this repo
|
||||
description: Work with Effect v4 / effect-smol TypeScript code in this repo
|
||||
---
|
||||
|
||||
# Effect
|
||||
@@ -9,10 +9,10 @@ This codebase uses Effect for typed, composable TypeScript services, schemas, an
|
||||
|
||||
## Source Of Truth
|
||||
|
||||
Use the current Effect v4 source, not memory or older Effect v2/v3 examples.
|
||||
Use the current Effect v4 / effect-smol source, not memory or older Effect v2/v3 examples.
|
||||
|
||||
1. If `.opencode/references/effect` is missing, clone `https://github.com/Effect-TS/effect` there. Do this in the project, not in the skill folder.
|
||||
2. Search `.opencode/references/effect` for exact APIs, examples, tests, and naming patterns before answering or implementing Effect-specific code.
|
||||
1. If `.opencode/references/effect-smol` is missing, clone `https://github.com/Effect-TS/effect-smol` there. Do this in the project, not in the skill folder.
|
||||
2. Search `.opencode/references/effect-smol` for exact APIs, examples, tests, and naming patterns before answering or implementing Effect-specific code.
|
||||
3. Also inspect existing repo code for local house style before introducing new patterns.
|
||||
4. Prefer answers and implementations backed by specific source files or nearby repo examples.
|
||||
|
||||
@@ -27,12 +27,12 @@ Use the current Effect v4 source, not memory or older Effect v2/v3 examples.
|
||||
- Keep layer composition explicit. Avoid broad hidden provisioning that makes missing dependencies hard to see.
|
||||
- In tests, prefer the repo's existing Effect test helpers and live tests for filesystem, git, child process, locks, or timing behavior.
|
||||
- Do not introduce `any`, non-null assertions, unchecked casts, or older Effect APIs just to satisfy types.
|
||||
- Do not answer from memory. Verify against `.opencode/references/effect` or nearby code first.
|
||||
- Do not answer from memory. Verify against `.opencode/references/effect-smol` or nearby code first.
|
||||
|
||||
## Testing Patterns
|
||||
|
||||
- Use `testEffect(...)` from `packages/core/test/lib/effect.ts` for tests that exercise Effect services, layers, runtime context, scoped resources, or platform integrations.
|
||||
- Use `testEffect(...)` from `packages/opencode/test/lib/effect.ts` for tests that exercise Effect services, layers, runtime context, scoped resources, or platform integrations.
|
||||
- Use `it.live(...)` for filesystem, git repositories, HTTP servers, sockets, child processes, locks, real time, and other live platform behavior.
|
||||
- Run tests from package directories such as `packages/core`; never run package tests from the repo root.
|
||||
- Run tests from package directories such as `packages/opencode`; never run package tests from the repo root.
|
||||
- Prefer explicit test layers over ad hoc managed runtimes. Keep dependency provisioning visible in the test file.
|
||||
- Use scoped fixtures and finalizers for resources that must be cleaned up, including temporary directories, flags, databases, fibers, servers, and global state.
|
||||
|
||||
@@ -1,63 +0,0 @@
|
||||
---
|
||||
name: rtl-aware-development
|
||||
description: OpenCode Desktop should be RTL-aware. Use when implementing or reviewing RTL/LTR behavior in the web app, desktop app, CSS, menus, scrolling, resizing, icons, mixed-direction text, or Electron title bars.
|
||||
---
|
||||
|
||||
# RTL-Aware Development
|
||||
|
||||
Treat direction as independent from language. Test English in both directions as well as real RTL and mixed-script content.
|
||||
|
||||
## Guidelines
|
||||
|
||||
- Set `lang` and `dir` on the document, and propagate direction through component providers used by portaled menus and popovers. Do not change the selected locale merely to force RTL.
|
||||
- Keep DOM and focus order semantic. Flexbox and Grid already follow `dir`; do not add `row-reverse`, CSS `order`, or reversed markup just to mirror a layout.
|
||||
- Prefer logical CSS for semantic layout. Reserve physical coordinates for pointer positions, canvas geometry, native window controls, and other genuinely physical placement.
|
||||
|
||||
```css
|
||||
/* Avoid */
|
||||
padding-left: 12px;
|
||||
right: 0;
|
||||
border-right: 1px solid;
|
||||
text-align: left;
|
||||
|
||||
/* Prefer */
|
||||
padding-inline-start: 12px;
|
||||
inset-inline-end: 0;
|
||||
border-inline-end: 1px solid;
|
||||
text-align: start;
|
||||
```
|
||||
|
||||
- Isolate mixed-direction text. Use `dir="auto"` or `<bdi>` for unknown text; keep code, URLs, IDs, and filesystem paths LTR without forcing the surrounding component LTR.
|
||||
|
||||
```html
|
||||
<span class="file-row"><bdi dir="auto">README.md</bdi></span> <bdi dir="ltr"><code>C:\src\app.ts</code></bdi>
|
||||
```
|
||||
|
||||
- Mirror directional meaning, not every image. Back/forward, previous/next, disclosure, indentation, and directional progress may need mirroring. Do not mirror brands, clocks, media controls, charts, or text. Reverse physical gradients, `translateX`, SVG transforms, and animation deltas explicitly.
|
||||
- Map interactions through direction. `clientX` remains physical; resizing a logical edge needs an RTL-aware delta. Logical previous/next keyboard controls may swap ArrowLeft/ArrowRight. Follow the relevant WAI-ARIA widget pattern.
|
||||
- Do not assume LTR scrolling. RTL `scrollLeft` can start at `0` and become negative. Prefer `scrollIntoView({ inline: "nearest" })` or a tested direction-normalizing helper.
|
||||
- For Electron title bars, prefer native caption controls and use `titleBarOverlay` plus `env(titlebar-area-*)` for the safe content rectangle. Keep Windows/macOS native-control avoidance and `trafficLightPosition` physical; keep app navigation inside that rectangle logical. Mark interactive titlebar children `app-region: no-drag`.
|
||||
- Verify behavior, not screenshots alone. Check computed styles, pseudo-element geometry, hit zones, focus order, keyboard behavior, submenu direction, zoom/scaling, and both LTR and RTL scroll endpoints.
|
||||
|
||||
## Test Matrix
|
||||
|
||||
- English + LTR
|
||||
- English + forced RTL
|
||||
- A real RTL locale + RTL
|
||||
- Mixed RTL/LTR content, long labels, numbers, code, and paths
|
||||
- Keyboard, pointer resize, scrolling, menus/submenus, and Electron titlebar controls in both directions
|
||||
|
||||
## References
|
||||
|
||||
- [RTL Styling 101, Ahmad Shadeed](https://rtlstyling.com/posts/rtl-styling/)
|
||||
- [CSS-Tricks: RTL Styling 101](https://css-tricks.com/rtl-styling-101/)
|
||||
- [CSS-Tricks: CSS Logical Properties and Values](https://css-tricks.com/css-logical-properties-and-values/)
|
||||
- [W3C: Structural markup and right-to-left text](https://www.w3.org/International/questions/qa-html-dir)
|
||||
- [W3C: Inline bidirectional markup](https://www.w3.org/International/articles/inline-bidi-markup/)
|
||||
- [MDN: CSS logical properties and values](https://developer.mozilla.org/en-US/docs/Web/CSS/Guides/Logical_properties_and_values)
|
||||
- [MDN: `dir`](https://developer.mozilla.org/en-US/docs/Web/HTML/Reference/Global_attributes/dir)
|
||||
- [MDN: `scrollLeft`](https://developer.mozilla.org/en-US/docs/Web/API/Element/scrollLeft)
|
||||
- [web.dev: Logical properties](https://web.dev/learn/css/logical-properties/)
|
||||
- [Electron: Custom title bar](https://www.electronjs.org/docs/latest/tutorial/custom-title-bar)
|
||||
- [WAI-ARIA: Window splitter pattern](https://www.w3.org/WAI/ARIA/apg/patterns/windowsplitter/)
|
||||
- [Kobalte: I18n Provider](https://kobalte.dev/docs/core/components/i18n-provider/)
|
||||
@@ -4,7 +4,7 @@ import { tool } from "@opencode-ai/plugin"
|
||||
const TEAM = {
|
||||
tui: ["kommander", "simonklee"],
|
||||
desktop_web: ["Hona", "Brendonovich"],
|
||||
core: ["jlongster", "rekram1-node", "neriousy", "nexxeln", "kitlangton"],
|
||||
core: ["jlongster", "rekram1-node", "nexxeln", "kitlangton", "starptech"],
|
||||
inference: ["fwang", "MrMushrooooom", "starptech"],
|
||||
windows: ["Hona"],
|
||||
} as const
|
||||
|
||||
@@ -1,30 +1,9 @@
|
||||
- After changing the public Protocol or Server `HttpApi`, run `bun run generate` from `packages/client`. Do not edit generated client files directly.
|
||||
- To regenerate the legacy JavaScript SDK, run `./packages/sdk/js/script/build.ts`.
|
||||
- After changing the public Protocol or Server `HttpApi`, run `bun run generate` from `packages/client`. Do not edit `src/generated` or `src/generated-effect` directly.
|
||||
- Keep runtime dependencies directed from Schema to Core and Protocol, then from Core and Protocol to Server. Client runtime code may depend on Schema and Protocol but never Core or Server; `sdk-next` composes Client, Core, and Server.
|
||||
- Current implementation changes belong in `packages/core`, `packages/cli`, `packages/server`, `packages/protocol`, `packages/schema`, and related generated client surfaces when required.
|
||||
- The default branch in this repo is `v2`.
|
||||
- Base all new branches and worktrees on `v2`, or `origin/v2` when the local `v2` ref is unavailable. Do not base them on `dev`.
|
||||
- Local `main` ref may not exist; use `v2` or `origin/v2` for diffs.
|
||||
|
||||
## Live V2 TUI Testing
|
||||
|
||||
- Run `bun run dev:live` from a development worktree to test its TUI against the currently elected `opencode2` background server and live sessions.
|
||||
- Pass a directory after the script when needed, for example `bun run dev:live /path/to/project`.
|
||||
- The script discovers the server with `opencode2 service status`, injects its private local credential from `opencode2 service get password`, and uses the `next` TUI storage channel so tabs and other client-local state match the installed client.
|
||||
- Prefer `dev:live` over plain `bun run dev` for this workflow. An implicit managed-service connection may replace the live server when the worktree client version differs; explicit `--server` warns and continues without replacing it.
|
||||
|
||||
## V2 TUI Stories
|
||||
|
||||
- When a user asks for a TUI story, add a fixture-driven story under `packages/tui/src/feature-plugins/system/storybook` and register it in `index.tsx`.
|
||||
- Render the real production component rather than a visual copy. Keep submissions and other side effects local to the story so it is safe to explore repeatedly.
|
||||
- Expose the meaningful state dimensions through story keybindings and list them in `StoryFooter`; include a reset command when combinations can leave the fixture in a confusing state.
|
||||
- Run a specific story with `OPENCODE_STORY=<story-id> bun run dev:live` from the development worktree, and exercise narrow and wide terminal sizes when layout is relevant.
|
||||
|
||||
## TUI Theme Tokens
|
||||
|
||||
- Choose theme tokens by semantic role, not by their current color. Do not use raw `theme.hue` values or borrow an unrelated semantic token to achieve a preferred appearance.
|
||||
- Use `text.feedback` and `background.feedback` only for outcome or status feedback such as errors, warnings, success messages, and informational messages. Use `formfield` states for form-control text, ordinals, and selection markers, and `action` states for actions.
|
||||
- If the theme does not expose a token for the required semantic role, extend the theme schema, defaults, resolution, and types with that role before using it in a component. Do not repurpose the nearest-looking existing token.
|
||||
- When changing the public theme token surface, verify the built-in light and dark defaults and the custom-theme fallback path in addition to the affected TUI component.
|
||||
- Do not modify `packages/opencode` unless the user explicitly asks for V1 work. `packages/opencode` is the V1 implementation and is present for reference only. New implementation changes should land in the V2 package set: `packages/core`, `packages/cli`, `packages/server`, `packages/protocol`, `packages/schema`, and related generated client surfaces when required.
|
||||
- The default branch in this repo is `dev`.
|
||||
- Local `main` ref may not exist; use `dev` or `origin/dev` for diffs.
|
||||
|
||||
## Branch Names
|
||||
|
||||
@@ -46,7 +25,6 @@ Examples: `fix(tui): simplify thinking toggle styling`, `docs: update contributi
|
||||
|
||||
- Keep things in one function unless composable or reusable
|
||||
- Do not extract single-use helpers preemptively. Inline the logic at the call site unless the helper is reused, hides a genuinely complex boundary, or has a clear independent name that improves the caller.
|
||||
- Before adding complexity for a speculative or vanishingly unlikely race or security edge case, explain the concrete failure mode, likelihood, and complexity cost to the user and get their buy-in. Do not silently expand scope for theoretical robustness.
|
||||
- Avoid `try`/`catch` where possible
|
||||
- Avoid using the `any` type
|
||||
- Use Bun APIs when possible, like `Bun.file()`
|
||||
@@ -82,7 +60,6 @@ const { a, b } = obj
|
||||
### Imports
|
||||
|
||||
- Never alias imports. Do not use `import { foo as bar } from "..."` or renamed imports like `resolve as pathResolve`.
|
||||
- Never use type-position `import("...")` references such as `Schema.declare<import("@opencode-ai/plugin/effect/plugin").Plugin["effect"]>`. Only when two imports genuinely collide on a name and no other option exists, an aliased type import (`import type { Plugin as PluginDefinition } from "..."`) is permitted as a last resort — still strongly preferred not to.
|
||||
- Never use star imports. Do not use `import * as Foo from "..."` or `import type * as Foo from "..."`.
|
||||
- If a namespace-style value is needed, import the module's own exported namespace by name, for example `import { Project } from "@opencode-ai/core/project"`, then reference `Project.ID`.
|
||||
- Prefer dynamic imports for heavy modules that are only needed in selected code paths, especially in startup-sensitive entrypoints. Destructure dynamic import bindings near the top of the narrowest scope that needs them so they read like normal imports. Avoid inline chains such as `await import("./module").then((mod) => mod.value())` or `(await import("./module")).value()`. Keep branch-specific imports inside the branch that needs them to preserve lazy loading.
|
||||
@@ -166,23 +143,23 @@ const table = sqliteTable("session", {
|
||||
|
||||
- Avoid mocks as much as possible, you shouldn't be using globalThis.\* at all unless it's the only option.
|
||||
- Test actual implementation, do not duplicate logic into tests
|
||||
- Tests cannot run from repo root (guard: `do-not-run-tests-from-root`); run from package directories such as `packages/core`.
|
||||
- Tests cannot run from repo root (guard: `do-not-run-tests-from-root`); run from package dirs like `packages/opencode`.
|
||||
|
||||
## Type Checking
|
||||
|
||||
- Always run `bun typecheck` from package directories (for example, `packages/core`), never `tsc` directly.
|
||||
- Always run `bun typecheck` from package directories (e.g., `packages/opencode`), never `tsc` directly.
|
||||
|
||||
## V2 Session Core
|
||||
|
||||
- Keep durable events minimal: record irreducible new facts and do not repeat state derivable by folding the ordered aggregate history. Enrich projections and read models with previous or derived state when consumers need self-contained views.
|
||||
- Keep durable prompt admission separate from model execution. `Session.prompt(...)` publishes `session.inbox.enqueued`, whose projection inserts one durable `session_inbox` row, before scheduling advisory `SessionExecution.wake(sessionID)` unless `resume: false` requests admit-only behavior. Delivery publishes `session.inbox.delivered`; its projection consumes the inbox row and inserts the visible message in the same transaction. `session_inbox` stores only unconsumed work.
|
||||
- Reusing a Session ID adopts the existing Session. While a user or synthetic inbox item is pending, reusing its ID reconciles only when Session, type, complete payload, metadata, and delivery match; conflicting reuse fails. Once delivered, retry reconciliation for those message-producing items uses the projected message and does not require retained enqueue history or the original delivery mode. Control items keep their operation-specific conflict behavior.
|
||||
- Keep `SessionExecution` process-global and Session-ID based. Its local implementation owns the process-local Session coordinator and discovers placement through `SessionStore` plus `LocationServiceMap.get(session.location)` only when a drain starts; no layer should take a Session ID. V2 interruption targets the active process-local ownership chain for that Session; interruption of a known but idle or locally unowned Session is a no-op, while the public API rejects an unknown Session.
|
||||
- Keep durable prompt admission separate from model execution. `SessionV2.prompt(...)` admits one durable `session_input` row before scheduling advisory `SessionExecution.wake(sessionID)` unless `resume: false` requests admit-only behavior. The serialized runner promotes admitted inputs into visible user messages at safe boundaries.
|
||||
- Reusing a Session ID adopts the existing Session. Reusing a prompt message ID reconciles an exact retry only when Session, prompt, and delivery mode match; conflicting reuse fails. Historical projected prompts lazily synthesize promoted inbox records during exact retry.
|
||||
- Keep `SessionExecution` process-global and Session-ID based. Its local implementation owns the process-local Session coordinator and discovers placement through `SessionStore` plus `LocationServiceMap.get(session.location)` only when a drain starts; no layer should take a Session ID. V2 interruption targets the active process-local ownership chain for that Session; idle or missing interruption is a no-op.
|
||||
- Keep `SessionRunner`, model resolution, tool registry, permissions, and filesystem Location-scoped. Omitted `Location.workspaceID` means implicit-local placement; explicit workspace identity remains reserved for future placement semantics.
|
||||
- Preserve one explicit `llm.stream(request)` call per Physical Attempt and reload projected history before durable continuation. A logical Step may use generic pre-output retries, one full-context retry after continuation rejection, incomplete-stream continuation, or one overflow-compaction rebuild. Generic retries retain the logical step number and do not consume another agent-step allowance. Do not delegate orchestration to an in-memory tool loop.
|
||||
- Keep local Session drains process-local until clustering is implemented. `SessionRunCoordinator` joins explicit same-Session resumes, coalesces prompt wakeups, and allows different Sessions to run concurrently. A write-ahead execution claim marks a process-local busy period for restart recovery: terminal completion, failure, or user interruption releases it, while shutdown interruption and process death preserve it. Startup recovery resumes claimed top-level Sessions with durable per-execution attempt accounting. The claim is a recovery marker, not clustered ownership, fencing, or an exactly-once guarantee.
|
||||
- Keep delivery vocabulary explicit. Prompts steer by default. Steers deliver in enqueue order at safe step boundaries, stopping before compaction or move control items. At an idle boundary, steers take priority; otherwise exactly one queued item delivers before the runner reevaluates continuation. Inbox items may be cancelled or changed between queue and steer before delivery. Promoting new user input resets the selected agent's step allowance; a batch of steers resets it once.
|
||||
- Preserve one explicit `llm.stream(request)` call per step and reload projected history before durable continuation. Do not bridge through legacy `SessionPrompt.loop(...)` or delegate orchestration to an in-memory tool loop.
|
||||
- Keep local Session drains process-local until clustering is implemented. `SessionRunCoordinator` joins explicit same-Session resumes, coalesces prompt wakeups, and allows different Sessions to run concurrently. Advisory wakes drain eligible durable inbox rows only; post-crash continuation recovery requires a separate explicit design before it may retry provider work. A drain has no durable identity or transcript boundary.
|
||||
- Keep delivery vocabulary explicit. Prompts steer by default and promote at the next safe step boundary while the current drain requires continuation. An explicit `queue` input remains pending until the Session would otherwise become idle; promote one queued input at that boundary, then reevaluate continuation before promoting another. Promoting any new user input resets the selected agent's step allowance; a batch of steers resets it once.
|
||||
- One step is one logical LLM call; its durable record covers only the model-visible span. Do not write "provider turn", and do not use bare "turn" for a single call: "turn" is reserved for the future assistant-turn unit containing all steps from prompt promotion until the session would go idle.
|
||||
- Keep event replay ownership separate from clustered Session execution ownership.
|
||||
- Keep the Instructions algebra and built-ins in `src/instructions`; keep instruction producers with their observed domains, and keep Session History selection plus `InstructionState` and `InstructionEntry` persistence Session-owned. `InstructionDiscovery` observes ambient global and upward-project instructions. The runner composes built-ins, discovery, guidance, and entries explicitly in `loadInstructions`; there is no instruction registry.
|
||||
- `session.instructions.updated` stores changed source keys and content hashes and may freeze rendered chronological update text. Blob values live once in `instruction_blob`; the projected `instruction_state` row is the normal boundary-processing source of current and initial values. Request assembly renders the epoch baseline from stored values, while later frozen updates enter history as durable System messages. Completed compaction moves the instruction epoch; Session movement retains it so destination instruction changes are chronological, while committed revert clears it. Forks adopt the parent's newest instruction values even when copied message history ends at an earlier boundary. Unavailable sources retain the last value and block only the initial complete delta.
|
||||
- Keep EventV2 replay owner claims separate from clustered Session execution ownership.
|
||||
- Keep the Instructions algebra and built-ins in `src/instructions`; keep instruction producers with their observed domains, and keep Session History selection plus `InstructionCheckpoint` and `InstructionEntry` persistence Session-owned. `InstructionDiscovery` observes ambient global and upward-project instructions. The runner composes built-ins, discovery, guidance, and entries explicitly in `loadInstructions`; there is no instruction registry.
|
||||
- The durable `Instructions.Applied` record is what the model was last told, per instruction source. Reconciliation narrates drift through `session.instructions.updated` and never rewrites the baseline; only completed compaction rebaselines, while Session movement or committed revert resets the `InstructionCheckpoint`. Unavailable sources keep the model's prior belief, blocking only a Session's first instruction baseline.
|
||||
|
||||
+245
@@ -0,0 +1,245 @@
|
||||
# OpenCode Session Runtime
|
||||
|
||||
OpenCode sessions preserve durable conversational history while assembling the runtime context an agent needs to act correctly in its current environment.
|
||||
|
||||
## Language
|
||||
|
||||
**Model Context**:
|
||||
The complete model-visible input assembled for one **Step**, including system instructions, **Session History**, tool definitions, and step-local additions. **Instructions** are one component of Model Context, not a synonym for it.
|
||||
_Avoid_: System Context
|
||||
|
||||
**Instructions**:
|
||||
The opaque algebra of independently refreshable typed instruction sources that render the durable instruction baseline and chronological updates shown to the model.
|
||||
_Avoid_: Model Context, System Context
|
||||
|
||||
**Session History**:
|
||||
The projected chronological conversation selected for a **Step** after applying the active compaction and **InstructionCheckpoint** baseline cutoffs.
|
||||
_Avoid_: Session Context
|
||||
|
||||
**Instruction Source**:
|
||||
One independently observed typed value within **Instructions**, represented by a stable namespaced key, JSON codec, loader, pure baseline/update renderers, and an optional removal renderer.
|
||||
_Avoid_: Prompt fragment
|
||||
|
||||
**InstructionEntry**:
|
||||
One API-managed, durable, per-Session instruction value. Its slash-free client key maps to the `api/<key>` **Instruction Source** key. Entries deliberately render to the model as mechanism-neutral `<context>` blocks: the model sees session context, not how it was attached.
|
||||
|
||||
**InstructionDiscovery**:
|
||||
The Location-scoped service that observes ambient global and upward-project `AGENTS.md` files as one ordered aggregate **Instruction Source**.
|
||||
|
||||
**InstructionCheckpoint**:
|
||||
The Session-owned durable instruction baseline, baseline sequence, and `Instructions.Applied` record used to prepare later Steps.
|
||||
|
||||
**Instruction Update**:
|
||||
A durable chronological System message published as `session.instructions.updated` that tells the model the newly effective state of one or more changed **Instruction Sources**.
|
||||
_Avoid_: System notification, raw text diff
|
||||
|
||||
**Instruction Baseline**:
|
||||
The exact joined instruction text stored by **InstructionCheckpoint** and sent as immutable provider-cache prefix state until completed compaction rebaselines it or Session movement or committed revert resets it.
|
||||
_Avoid_: Live system prompt
|
||||
|
||||
**Applied Instructions**:
|
||||
The overwriteable model-hidden `Instructions.Applied` record in **InstructionCheckpoint**, containing what the model was last told per **Instruction Source**.
|
||||
|
||||
**Unavailable Instruction Source**:
|
||||
An expected temporary inability to observe an **Instruction Source** value; the runtime retains its prior effective state and emits no update, while an unavailable source blocks creation of the first complete **Instruction Baseline**.
|
||||
|
||||
**Safe Step Boundary**:
|
||||
The point during Step preparation, after prior tool settlement and before durable input promotion, where instruction changes may be admitted chronologically.
|
||||
|
||||
**Admitted Prompt**:
|
||||
A durable user input accepted into the Session inbox but not yet included in **Session History**.
|
||||
|
||||
**Prompt Promotion**:
|
||||
The durable transition that removes an **Admitted Prompt** from pending input and appends its user message to **Session History**.
|
||||
|
||||
**Step**:
|
||||
One logical LLM call spanning pre-flight instruction checkpoint preparation, input promotion, request build, and compaction check; the provider stream; and tool settlement.
|
||||
_Avoid_: provider turn, turn (unqualified)
|
||||
|
||||
**Physical Attempt**:
|
||||
One actual provider request on the wire in service of a **Step**; most Steps have one Physical Attempt, while overflow-triggered compaction recovery may give one Step two.
|
||||
|
||||
**Assistant Turn**:
|
||||
A reserved name for the not-yet-modeled unit containing all **Steps** from prompt promotion until the assistant yields the floor; do not reify it until something durable needs it.
|
||||
|
||||
**Settlement**:
|
||||
The terminal transition for a unit of work: Step and tool settlement are durable, while drain and execution settlement are coordinator-observed.
|
||||
|
||||
**Execution**:
|
||||
One session-scoped coordinator busy period from first wake until idle. An Execution is process-local coordination rather than a durable domain entity.
|
||||
|
||||
**Session Drain**:
|
||||
One process-local execution span that promotes eligible input and runs required **Steps** until no immediate continuation remains. A Session Drain has no durable identity or transcript boundary.
|
||||
|
||||
**Model Tool Output**:
|
||||
The bounded projection of a Core-executed tool result persisted in Session history and replayed to the model. A tool may shape this projection semantically, but the Tool Registry enforces the final size limit.
|
||||
|
||||
**Managed Tool Output File**:
|
||||
A temporary file created under OpenCode's shared tool-output directory to retain complete output that was too large for Session history.
|
||||
|
||||
**Model Request Options**:
|
||||
Provider-semantic model settings selected from the Catalog and active Session variant before the LLM protocol adapter encodes them for a provider request.
|
||||
_Avoid_: Request body, wire options
|
||||
|
||||
**Generation Controls**:
|
||||
Provider-neutral sampling and output controls, partitioned from provider semantics and compatibility wire fields when model metadata enters the Catalog.
|
||||
|
||||
**Native Continuation Metadata**:
|
||||
Opaque protocol-shaped data attached to assistant content and required to continue that content natively with a compatible model, such as a reasoning signature or provider-hosted item identifier.
|
||||
|
||||
**PTY Environment**:
|
||||
The host-supplied environment overlay applied by the server when creating a PTY, observed for the request Location and resolved PTY working directory.
|
||||
|
||||
**OpenCode Client**:
|
||||
The generated Promise and Effect APIs derived from the public `HttpApi`; **Embedded OpenCode** shares the Effect API through an in-memory `HttpClient` against the same router and handlers.
|
||||
_Avoid_: Remote client
|
||||
|
||||
**SDK Contract IR**:
|
||||
The runtime-neutral compiled representation of the authoritative `HttpApi`, preserving encoded and decoded type projections plus transport metadata so independent SDK emitters can choose their public value model and runtime interpreter.
|
||||
|
||||
**Embedded OpenCode**:
|
||||
A scoped in-process host that structurally extends the **OpenCode Client**, supplies an in-memory HTTP transport, and exposes additional same-process capabilities directly.
|
||||
_Avoid_: Local implementation
|
||||
|
||||
**Page**:
|
||||
A bounded ordered result containing `items` and opaque `previous` and `next` cursor links for navigating the same query in either direction.
|
||||
_Avoid_: Response envelope
|
||||
|
||||
## Relationships
|
||||
|
||||
- **Instructions** is an opaque carrier composed from zero or more **Instruction Sources**.
|
||||
- **Model Context** is broader than **Instructions**. For each **Step**, the runner assembles the selected agent or provider system text, the **Instruction Baseline**, **Session History**, available tools, and step-local additions into one model request.
|
||||
- **Session History** contains projected conversational messages and admitted **Instruction Updates**; the active **Instruction Baseline** remains separate provider-request state.
|
||||
- The runner explicitly loads and combines instruction built-ins, **InstructionDiscovery**, selected-agent skill guidance, reference guidance, MCP guidance, and **InstructionEntry** values. There is no instruction registry.
|
||||
- `Instructions.combine(...)` preserves caller order and rejects duplicate stable namespaced source keys. The runner loads its producers concurrently, then combines them in its fixed declared order.
|
||||
- Each **Instruction Source** loader returns one coherent typed value or explicitly reports unavailability. `Instructions.make(...)` hides the value type so differently typed sources compose uniformly; its codec compares and stores the value, while pure renderers produce baseline, update, and optional removal text.
|
||||
- `Instructions.initialize(...)` observes composed **Instructions** once and produces a complete **Instruction Baseline** with **Applied Instructions**.
|
||||
- `Instructions.reconcile(...)` observes composed **Instructions** once and returns either unchanged or one combined chronological update. It never rewrites the baseline.
|
||||
- `Instructions.rebaseline(...)` renders a fresh baseline after completed compaction, recalling previously applied values for sources that are temporarily unavailable.
|
||||
- A changed **Instruction Source** may contribute text to one **Instruction Update** containing the newly effective state.
|
||||
- An **Instruction Update** persists the exact combined rendered text sent to the model through `session.instructions.updated`.
|
||||
- **Applied Instructions** advances atomically with the corresponding durable **Instruction Update**.
|
||||
- **Applied Instructions** stores one codec-encoded JSON value and, for removable sources, a pre-rendered removal message per stable **Instruction Source** key.
|
||||
- Changes from multiple **Instruction Sources** admitted at one safe boundary combine into one **Instruction Update**.
|
||||
- Instruction changes are sampled and admitted lazily at a **Safe Step Boundary**, never pushed asynchronously when their source changes.
|
||||
- At a **Safe Step Boundary**, prior tool results are already settled; instruction preparation completes before newly admitted user input promotes.
|
||||
- An **Admitted Prompt** is replayable pending input, not yet model-visible **Session History**.
|
||||
- **Prompt Promotion** atomically consumes the pending inbox entry and appends its model-visible user message.
|
||||
- Steering prompts promote at the next **Safe Step Boundary** while the current **Session Drain** still requires continuation. Promoting any newly admitted user input resets the selected agent's step allowance; multiple prompts promoted at one boundary reset it once.
|
||||
- A queued prompt does not promote while the current **Session Drain** requires continuation. The runner promotes one queued prompt when the Session would otherwise become idle, then reevaluates continuation before promoting another.
|
||||
- A **Session Drain** is process-local coordination rather than a durable domain entity. Durable recovery must reason from prompts, projected history, physical attempts, and tool state rather than inventing an enclosing execution identity.
|
||||
- An **Execution** contains one or more **Session Drains**; a **Session Drain** contains one reserved assistant-turn span at a time; that span contains **Steps**; and each **Step** contains one or more **Physical Attempts** plus any tool calls it requires.
|
||||
- A **Step** record covers only the model-visible span from first assistant output through tool settlement; pre-flight leaves no record, and one Step settles at most one record.
|
||||
- The first **Step** renders the latest complete **Instruction Baseline** and creates its **InstructionCheckpoint** without emitting a redundant **Instruction Update**; an unavailable initial source blocks the Step instead of persisting an incomplete baseline.
|
||||
- Instruction preparation precedes durable input promotion on every Step so an unavailable first baseline leaves pending input untouched and later updates enter history before newly promoted input.
|
||||
- Completed compaction rebaselines the **InstructionCheckpoint** from current **Instructions** and removes earlier **Instruction Updates** from active projected model history while preserving durable audit history.
|
||||
- A newly composed **Instruction Source** absent from **Applied Instructions** emits its baseline rendering once at the next **Safe Step Boundary**.
|
||||
- **Unavailable Instruction Source** uses stale-while-revalidate semantics and is distinct from a successfully loaded absence, which may emit removal text.
|
||||
- **InstructionDiscovery** observes ambient instructions as one ordered aggregate **Instruction Source**.
|
||||
- Ambient discovery reads global and upward-project `AGENTS.md` files and honors `OPENCODE_DISABLE_PROJECT_CONFIG` for project files.
|
||||
- After a successful internal file or directory read, nearby `AGENTS.md` files toward the Location root are injected once per Session as durable synthetic instruction messages.
|
||||
- **InstructionEntry** stores API-managed per-Session JSON values. Each entry contributes one `api/<key>` **Instruction Source**, so adding, replacing, or removing an entry is reconciled at the next **Safe Step Boundary**.
|
||||
- Location-scoped instruction producers naturally re-resolve when a moved Session next runs in its destination Location.
|
||||
- Moving a Session resets its **InstructionCheckpoint**, so the destination must initialize a complete baseline before another prompt can promote. Committed revert also resets the checkpoint.
|
||||
- Selected-agent available-skill guidance is an **Instruction Source** composed explicitly by the runner. It lists only names and descriptions permitted for that agent; skill bodies and locations are exposed only through the permission-checked `skill` tool.
|
||||
- The selected agent and model are sampled when a **Step** starts. Changes admitted after that boundary apply to the next Step and do not restart the current Step.
|
||||
- An agent switch that changes selected-agent guidance produces an **Instruction Update** while preserving the current baseline.
|
||||
- Local tool authorization and pending permission requests retain the effective agent of the **Step** that issued the call; a later agent switch cannot change that call's policy.
|
||||
- Instruction source changes never wake idle Sessions; the next naturally scheduled **Safe Step Boundary** loads and compares current values lazily.
|
||||
- Once admitted, an **Instruction Update** remains durable even if the following **Physical Attempt** fails and is replayed unchanged on retry.
|
||||
- **Instruction Updates** remain durable Session-message history; normal user-facing transcript surfaces may hide them.
|
||||
- The date **Instruction Source** initially preserves host-local calendar-date behavior; a configured user timezone may replace that default later.
|
||||
- An **Instruction Baseline** is stored durably and reused verbatim across process restarts until rebaseline or reset.
|
||||
- An **Instruction Baseline** durably preserves the exact joined text used for its part of the active provider-cache prefix.
|
||||
- A model/provider switch preserves the current **InstructionCheckpoint** and chronological conversation history; the new selection applies to the next **Step**.
|
||||
- **Native Continuation Metadata** remains in durable history. Step projection includes it only for a successful exact originating provider/model match; failed Steps and incompatible models omit opaque metadata, while non-empty visible reasoning lowers to ordinary assistant text after a model switch. This conservative relation may widen only when recorded provider tests establish compatibility.
|
||||
- **Model Request Options** remain provider-semantic through Catalog resolution. The Session runner maps them into the LLM package's provider-option namespace; the selected protocol adapter alone owns provider wire encoding.
|
||||
- **Generation Controls**, protocol-semantic **Model Request Options**, and compatibility request body fields are separate Catalog domains. A shared ingestion adapter partitions legacy and models.dev AI-SDK-shaped options before routing.
|
||||
- The **PTY Environment** is a server concern rather than a Core PTY concern. PTY creation merges caller values, then the host overlay, then Core-forced terminal invariants such as `TERM` and `OPENCODE_TERMINAL`.
|
||||
- Networked and **Embedded OpenCode** use the same **OpenCode Client** and preserve the full HTTP encoding, routing, middleware, and decoding boundary; only the `HttpClient` transport differs.
|
||||
- The Effect-native network constructor obtains `HttpClient.HttpClient` from its environment so callers own transport selection, recording, tracing, retries, and tests. Convenience runtimes may provide a fetch transport separately.
|
||||
- Creating **Embedded OpenCode** is scoped. Closing its owning Scope releases the in-process server resources, database resources, registrations, and fibers.
|
||||
- **Embedded OpenCode** exposes shared client capabilities and embedded-only capabilities on one object; consumers do not navigate through a nested `.client` property.
|
||||
- The beta **OpenCode Client** currently uses plural consumer-facing capability groups such as `sessions`; whether the stable Session namespace should instead be singular `session` must be settled before stabilization. Internal server identifiers do not implicitly define public client names.
|
||||
- Server's concrete `HttpApi` is authoritative for shared **OpenCode Client** capabilities. Codegen compiles its Session group directly; the Effect runtime uses an equivalent Protocol-only projection so generated artifacts remain independent of Core and Server.
|
||||
- SDK generation reflects the public `HttpApi` once into an **SDK Contract IR**. Promise and Effect emitters share endpoint structure and transport metadata without being required to expose identical public values: an emitter may select encoded wire types, decoded domain types, compile-time brands, runtime validation, and its own execution abstraction independently.
|
||||
- The first Effect emitter is the rich projection: it exposes decoded Effect-native values, preserves brands and schema transformations, performs runtime schema decoding, and delegates transport interpretation to `HttpApiClient`. Lighter wire-shaped Effect output remains possible through another emitter policy rather than constraining the shared IR.
|
||||
- The rich Effect emitter regenerates private executable schemas when the **SDK Contract IR** proves that their transport semantics can be reproduced exactly. Contracts with authoritative custom transformations use the import-based Effect emitter against a Protocol-only client projection whose generated transport output is tested against Server's concrete API; the Promise emitter still derives zero-Effect structural wire types from the same IR.
|
||||
- `@opencode-ai/protocol` owns Session endpoint construction and middleware placement. Server supplies concrete middleware keys to produce the authoritative build-time API; the client projection supplies transport-only keys without importing Core or Server at runtime.
|
||||
- The first Promise emitter targets the same clean domain-oriented method organization rather than Hey API source compatibility. It returns unwrapped values directly, rejects declared and infrastructure failures, and begins with minimal client-level transport configuration; result wrappers, interceptors, and legacy generated signatures are outside the initial surface.
|
||||
- The first Promise emitter parses response syntax and trusts its generated structural types; it does not perform runtime structural validation. Malformed payload syntax fails, while a syntactically valid shape mismatch is not detected at the SDK boundary. Standalone validator generation remains an optional future emitter policy.
|
||||
- Declared Promise-client failures retain their tagged structural wire values and have generated type guards. Consumers do not depend on generated `Error` subclass identity, preserving discrimination across package copies and realms while remaining structurally aligned with Effect domain errors.
|
||||
- Promise-client infrastructure failures use one generated `ClientError` class with a structured reason such as transport failure, unexpected status, unsupported content type, or malformed response. Promise methods reject with either a tagged declared domain failure or `ClientError`, matching the Effect client's conceptual domain/infrastructure error division.
|
||||
- Promise methods accept a separate optional per-call transport-options argument containing `AbortSignal` and header overrides. Cancellation and transport metadata do not enter the domain input object; broader interceptor and response-mode APIs remain deferred.
|
||||
- Promise streaming methods return a lazy `AsyncIterable` directly rather than a Promise-wrapped stream object. Iteration opens the connection, `AbortSignal` cancels it, and ending iteration closes the underlying request; the Effect emitter analogously returns `Stream` directly.
|
||||
- Promise SSE connection establishment, declared HTTP failures, and infrastructure failures occur during `AsyncIterable` iteration, beginning with its first `next()` call, rather than during synchronous method construction.
|
||||
- Neither generated streaming runtime automatically reconnects after disconnection. Promise `AsyncIterable` and Effect `Stream` fail explicitly; live consumers refresh and resubscribe, while durable sequence-based resume remains explicit composition above the generated client.
|
||||
- Promise client construction is synchronous and network-free. It requires `baseUrl`, defaults to `globalThis.fetch`, accepts client-level headers, and merges them with per-call header overrides.
|
||||
- Effect client construction accepts an explicit `baseUrl` and obtains `HttpClient.HttpClient` from the Effect environment. It does not install fetch or duplicate per-call transport policy; callers transform/provide the client for headers, tracing, retries, recording, and tests, while fiber interruption owns cancellation.
|
||||
- Promise and Effect emitters each own their generated public type modules. The **SDK Contract IR**, not a physically shared generated type package, is the common source; this permits zero-Effect wire types and rich decoded Effect types to evolve independently.
|
||||
- Promise and Effect network clients ship from `@opencode-ai/client` behind isolated root and `/effect` exports. The root has no runtime path to Effect; `/effect` imports only Effect, Schema, and Protocol.
|
||||
- The Effect-native scoped host belongs to `@opencode-ai/sdk-next`, which will assume the existing `@opencode-ai/sdk` name after legacy consumers migrate. Client remains network-only and SDK depends one-way on Client.
|
||||
- SDK executes Server's assembled `HttpRouter` in memory. It opens no listener and performs no network I/O, while preserving Server routing, middleware, codecs, handlers, and errors.
|
||||
- The Effect Client and SDK re-export their decoded datatype facade from Schema so callers do not depend on internal package locations or Core's versioned names.
|
||||
- A capability intended for both networked and **Embedded OpenCode** belongs in the authoritative public `HttpApi`; embedded-only same-process capabilities extend **Embedded OpenCode** separately.
|
||||
- `sessions.events({ sessionID, after })` is a public durable Session event stream. It verifies the Session, replays durable events after the optional aggregate sequence, continues with newly committed durable events, excludes live-only fragments, and is transported as SSE in both networked and embedded modes.
|
||||
- `events.subscribe()` is a distinct public instance-wide live stream for Session and non-Session activity. It has no replay guarantee and includes connection, heartbeat, and instance-disposal lifecycle events; consumers recover from disconnection by refreshing authoritative state.
|
||||
- A Session ID is not an optional filter on `events.subscribe()`: instance-wide live events and durable Session events have different schemas, replay guarantees, cursors, lifecycle events, and failure behavior.
|
||||
- The initial common OpenCode Client does not expose server-global event aggregation. `events.subscribe()` is bounded to the connected OpenCode instance or workspace; any future cross-instance administrative stream requires a separately designed API.
|
||||
- `events.subscribe()` does not automatically reconnect after transport loss. The live-only stream fails with `ClientError`; consumers refresh authoritative state before explicitly opening a new subscription because events missed during disconnection cannot be replayed.
|
||||
- `sessions.events({ sessionID, after })` returns the generated HTTP client's cold durable event stream and does not build reconnection policy into the endpoint or client constructor. Transport loss fails the stream with `ClientError`. Callers may compose an explicit resuming stream above it by retaining the last observed durable sequence and opening a new subscription with `after`; any reusable resume helper remains a separate API design question.
|
||||
- The stable `sessions.list(...)` design returns a **Page** in both networked and **Embedded OpenCode**; embedded execution does not define a separate unbounded array-returning list operation. The beta client currently preserves the existing HTTP `{ data, cursor }` envelope until emitter-level Page projection is implemented.
|
||||
- Session list cursors are opaque branded values carrying continuation query and ordering state. Consumers pass them back unchanged and do not inspect storage anchors or encoded filter fields.
|
||||
- A Session list continuation accepts only its opaque cursor. Scope, filters, ordering, and page size are fixed by the initial query and carried by that cursor.
|
||||
- `sessions.messages(...)` returns a **Page** and uses the same cursor discipline as `sessions.list(...)`: the initial request supplies `sessionID`, ordering, and page size; continuation supplies `sessionID` plus only an opaque branded message cursor carrying ordering, page size, direction, and message anchor. Using a cursor with another Session is invalid.
|
||||
- `sessions.message({ sessionID, messageID })` is a required resource lookup. An unknown Session fails with `SessionNotFoundError`; a known Session with an absent or differently owned message fails with `MessageNotFoundError` without disclosing cross-Session ownership. Absence is not represented as `undefined` across the public HTTP boundary.
|
||||
- `sessions.interrupt({ sessionID })` first verifies that the durable Session exists, failing with `SessionNotFoundError` otherwise. For a known Session, interruption is idempotent: idle, already-settled, or locally unowned execution is a no-op.
|
||||
- `sessions.active()` snapshots the current process's foreground Session drain registry as a record of Session IDs to `{ type: "running" }`. Missing IDs are inactive; background subagents and tasks do not make their parent Session active, and process restart clears the registry.
|
||||
- `sessions.context({ sessionID })` preserves the existing message-only operation. It returns projected **Session History**; it does not include or represent the complete **Model Context**, whose system text, **Instruction Baseline**, tools, and step-local additions remain separate.
|
||||
- **Open question**: Should a future, separately named operation expose complete **Model Context**, including the instruction baseline, applied instruction metadata, tools, and step-local additions?
|
||||
- `sessions.prompt(...)` exposes `resume?: boolean`. Omitting it preserves durable admission followed by an advisory execution wake; `resume: false` requests durable admit-only behavior.
|
||||
- The public operation remains `sessions.prompt(...)`; `SessionInput.admit` is the internal primitive, while the public `Admission` result and `resume` option express its durable admission semantics.
|
||||
- `sessions.create(...)` accepts an optional `location`. Omission resolves through the connected OpenCode instance's default or current location; an explicit value selects a known location. Networked and embedded transports use the same handler semantics.
|
||||
- `sessions.switchAgent({ sessionID, agent })` is part of the common client alongside `sessions.switchModel(...)`. It affects subsequent Session activity and fails with `SessionNotFoundError` for an unknown Session.
|
||||
- The **Embedded OpenCode** Layer delegates to the same scoped creation path; it does not define a second implementation.
|
||||
- A **PTY Environment** adapter observes plugins in the request Location while passing the resolved PTY working directory to the hook; standalone servers use an empty adapter.
|
||||
- An **Instruction Update** lowers to the provider's native chronological instruction role when supported and to a wrapped chronological fallback otherwise.
|
||||
- When the aggregate discovered instruction set changes, its **Instruction Update** includes the complete current ordered set and supersedes the prior aggregate value; when no discovered instructions remain, the message states that previously loaded instructions no longer apply.
|
||||
- Ambient project instruction discovery honors `OPENCODE_DISABLE_PROJECT_CONFIG`; global instructions remain eligible.
|
||||
- Oversized textual **Model Tool Output** retains a bounded preview in Session history while its complete text moves to managed tool-output storage. Arbitrary structured-result size is a separate concern.
|
||||
- One tool settlement receives one aggregate textual limit, using the configured maximum lines or UTF-8 bytes, whichever is reached first. The limit is provider-independent; token pressure belongs to context assembly and compaction.
|
||||
- Generic truncation preserves the beginning and end of textual output. Tools may apply a more meaningful strategy before the Tool Registry enforces the final limit.
|
||||
- A truncated **Model Tool Output** identifies its complete text both in the bounded model-visible preview and as a typed managed output path. Managed output paths do not modify the tool's validated structured result.
|
||||
- A **Managed Tool Output File** is temporary and may expire after its retention period. The bounded **Model Tool Output**, not the file, is the durable replayable record.
|
||||
- Failure to retain a **Managed Tool Output File** does not change a successful tool operation into a failed one. The Session records an explicitly lossy bounded output without a path, while operators receive diagnostics for the storage failure.
|
||||
- Once a tool operation succeeds, bounding its **Model Tool Output** and publishing its one durable settlement form an interruption-safe completion region. Raw oversized success is never published before a later correction.
|
||||
- When a structured-only result would exceed the **Model Tool Output** limit, its validated structured value remains unchanged for Session consumers while model replay uses a bounded textual JSON preview and optional managed output path.
|
||||
- Existing tool-managed output paths survive generic bounding. A fallback file retains exactly the complete projected text received by the Tool Registry and never claims to reconstruct output already discarded by tool-specific shaping.
|
||||
- **Managed Tool Output Files** use globally unique names in one shared flat directory. Their absolute paths are readable and searchable by ordinary tools; other absolute paths remain outside Location-scoped filesystem authority.
|
||||
- Provider-executed tool results remain provider-native transcript facts outside generic Tool Registry bounding. Their context control requires provider-aware pruning or compaction because some providers require exact structured round-trip payloads.
|
||||
|
||||
## Client contract architecture
|
||||
|
||||
Semantic values that mean the same thing internally and publicly live in the lightweight Schema leaf. Core consumes Schema for domain behavior; Protocol composes Schema values into paths, payloads, envelopes, errors, cursors, and streams; Server imports both, hosts Protocol's exact groups, and owns protocol/domain adaptation. The root Promise client remains zero-Effect, `/effect` depends on Effect plus Schema and Protocol, and `@opencode-ai/sdk-next` composes the scoped in-process host above Client, Core, and Server.
|
||||
|
||||
Shared public records are plain objects declared with `Schema.Struct`. A same-name inferred interface gives object records readable TypeScript signatures without constructors, prototypes, or nominal identity; unions retain explicit type aliases.
|
||||
|
||||
Before stabilizing the client API:
|
||||
|
||||
- Keep additional public schemas in Schema and additional network groups in Protocol; neither package may transitively load databases, Drizzle, Session execution, providers, watchers, native modules, or WASM.
|
||||
- Keep concrete Location middleware keys in Server while Protocol owns their placement. Client projections may supply transport-only keys, but must prove generated equivalence with Server's concrete API.
|
||||
- Project the existing list response envelope to the stable client **Page** shape and enforce separate initial-query and cursor-continuation inputs without changing the hosted V2 wire contract.
|
||||
- Settle the stable consumer namespace (`session` versus the current beta `sessions`) and use an explicit codegen annotation if the consumer name should differ from the server group identifier.
|
||||
- Preserve V2 route paths, operation IDs, codecs, errors, middleware behavior, and OpenAPI output while making this change.
|
||||
- Preserve browser-safe `@opencode-ai/client` and `@opencode-ai/client/effect` bundles through import-boundary tests.
|
||||
- Define embedded-host placement before supporting multiple hosts over one database. Hosts that share durable Session storage must also share process-local Session execution coordination, or each host must receive isolated storage explicitly.
|
||||
- Keep an embedded request scope alive until any streamed response body finishes. The initial non-streaming Session surface does not exercise this lifetime boundary; Session and instance event streams must do so before joining the embedded client.
|
||||
|
||||
## Example dialogue
|
||||
|
||||
> **Dev:** "The date changed while the session was active. Should the **Instruction Update** say what the old date was?"
|
||||
> **Domain expert:** "No. Emit the newly effective date so the agent can act on the current instructions."
|
||||
|
||||
## Flagged ambiguities
|
||||
|
||||
- Legacy `experimental.chat.system.transform` can mutate assembled system text arbitrarily, but V2 plugins do not yet expose an equivalent hook. Decide separately whether to port it, model dynamic uses as explicit **Instruction Sources**, or narrow its semantics.
|
||||
+250
-63
@@ -1,112 +1,299 @@
|
||||
# Contributing to OpenCode
|
||||
|
||||
The changes most likely to be accepted are:
|
||||
We want to make it easy for you to contribute to OpenCode. Here are the most common type of changes that get merged:
|
||||
|
||||
- Bug fixes
|
||||
- Additional LSPs and formatters
|
||||
- LLM performance improvements
|
||||
- Environment-specific fixes
|
||||
- Additional LSPs / Formatters
|
||||
- Improvements to LLM performance
|
||||
- Support for new providers
|
||||
- Fixes for environment-specific quirks
|
||||
- Missing standard behavior
|
||||
- Documentation improvements
|
||||
|
||||
UI and core product features require design review before implementation. If you are unsure whether a change fits, ask a maintainer or choose an issue labeled [`help wanted`](https://github.com/anomalyco/opencode/issues?q=is%3Aissue%20state%3Aopen%20label%3Ahelp-wanted), [`good first issue`](https://github.com/anomalyco/opencode/issues?q=is%3Aissue%20state%3Aopen%20label%3A%22good%20first%20issue%22), [`bug`](https://github.com/anomalyco/opencode/issues?q=is%3Aissue%20state%3Aopen%20label%3Abug), or [`perf`](https://github.com/anomalyco/opencode/issues?q=is%3Aopen%20is%3Aissue%20label%3A%22perf%22).
|
||||
However, any UI or core product feature must go through a design review with the core team before implementation.
|
||||
|
||||
Want to take on an issue? Leave a comment and a maintainer may assign it unless it is already being worked on.
|
||||
If you are unsure if a PR would be accepted, feel free to ask a maintainer or look for issues with any of the following labels:
|
||||
|
||||
- [`help wanted`](https://github.com/anomalyco/opencode/issues?q=is%3Aissue%20state%3Aopen%20label%3Ahelp-wanted)
|
||||
- [`good first issue`](https://github.com/anomalyco/opencode/issues?q=is%3Aissue%20state%3Aopen%20label%3A%22good%20first%20issue%22)
|
||||
- [`bug`](https://github.com/anomalyco/opencode/issues?q=is%3Aissue%20state%3Aopen%20label%3Abug)
|
||||
- [`perf`](https://github.com/anomalyco/opencode/issues?q=is%3Aopen%20is%3Aissue%20label%3A%22perf%22)
|
||||
|
||||
> [!NOTE]
|
||||
> PRs that ignore these guardrails will likely be closed.
|
||||
|
||||
## Adding Providers
|
||||
Want to take on an issue? Leave a comment and a maintainer may assign it to you unless it is something we are already working on.
|
||||
|
||||
New providers should rarely require OpenCode changes. Add the provider to [models.dev](https://github.com/anomalyco/models.dev) first.
|
||||
## Adding New Providers
|
||||
|
||||
## Development
|
||||
New providers shouldn't require many if ANY code changes, but if you want to add support for a new provider first make a PR to:
|
||||
https://github.com/anomalyco/models.dev
|
||||
|
||||
OpenCode requires Bun 1.3 or newer. From the repository root:
|
||||
## Developing OpenCode
|
||||
|
||||
- Requirements: Bun 1.3+
|
||||
- Install dependencies and start the dev server from the repo root:
|
||||
|
||||
```bash
|
||||
bun install
|
||||
bun dev
|
||||
```
|
||||
|
||||
### Running against a different directory
|
||||
|
||||
By default, `bun dev` runs OpenCode in the `packages/opencode` directory. To run it against a different directory or repository:
|
||||
|
||||
```bash
|
||||
bun install
|
||||
bun dev [directory]
|
||||
bun dev <directory>
|
||||
```
|
||||
|
||||
`bun dev` runs the V2 CLI and TUI. Pass a directory to open another project, or `.` to open this repository.
|
||||
|
||||
To test a development TUI against your installed OpenCode V2 background service and live sessions:
|
||||
To run OpenCode in the root of the opencode repo itself:
|
||||
|
||||
```bash
|
||||
bun run dev:live [directory]
|
||||
bun dev .
|
||||
```
|
||||
|
||||
For web development, run the backend and app in separate terminals. Other interfaces have root scripts:
|
||||
### Building a "localcode"
|
||||
|
||||
To compile a standalone executable:
|
||||
|
||||
```bash
|
||||
bun dev serve --port 4096
|
||||
bun run dev:web
|
||||
bun run dev:desktop
|
||||
bun run dev:www
|
||||
./packages/opencode/script/build.ts --single
|
||||
```
|
||||
|
||||
### Packages
|
||||
|
||||
- `packages/schema`: shared wire and storage contracts
|
||||
- `packages/core`: domain behavior and persistence
|
||||
- `packages/protocol`: public API definitions
|
||||
- `packages/server`: HTTP server and runtime composition
|
||||
- `packages/client`: generated TypeScript clients
|
||||
- `packages/cli`: command-line entrypoint and service lifecycle
|
||||
- `packages/tui`: terminal interface
|
||||
- `packages/app`: shared web interface
|
||||
- `packages/desktop`: Electron desktop application
|
||||
- `packages/plugin`: plugin API
|
||||
|
||||
### Verification
|
||||
|
||||
Run typechecks, and tests where defined, from the affected package rather than the repository root:
|
||||
Then run it with:
|
||||
|
||||
```bash
|
||||
cd packages/core
|
||||
bun run test
|
||||
bun typecheck
|
||||
./packages/opencode/dist/opencode-<platform>/bin/opencode
|
||||
```
|
||||
|
||||
Follow package-specific instructions in nearby `AGENTS.md` files. After changing the public Protocol or Server `HttpApi`, run `bun run generate` from `packages/client`; never edit generated client files directly.
|
||||
Replace `<platform>` with your platform (e.g., `darwin-arm64`, `linux-x64`).
|
||||
|
||||
Follow the repository [style guide](./AGENTS.md).
|
||||
- Core pieces:
|
||||
- `packages/opencode`: OpenCode core business logic & server.
|
||||
- `packages/opencode/src/cli/cmd/tui/`: The TUI code, written in SolidJS with [opentui](https://github.com/sst/opentui)
|
||||
- `packages/app`: The shared web UI components, written in SolidJS
|
||||
- `packages/desktop`: The native desktop app, built with Electron (wraps `packages/app`)
|
||||
- `packages/plugin`: Source for `@opencode-ai/plugin`
|
||||
|
||||
## Pull Requests
|
||||
### Understanding bun dev vs opencode
|
||||
|
||||
### Link Issues When Required
|
||||
During development, `bun dev` is the local equivalent of the built `opencode` command. Both run the same CLI interface:
|
||||
|
||||
Bug fixes, chores, and tests must reference an existing issue. Documentation, refactor, and feature PRs are exempt from the automated linked-issue check. When required, use `Fixes #123` or `Closes #123` in the PR description.
|
||||
```bash
|
||||
# Development (from project root)
|
||||
bun dev --help # Show all available commands
|
||||
bun dev serve # Start headless API server
|
||||
bun dev web # Start server + open web interface
|
||||
bun dev <directory> # Start TUI in specific directory
|
||||
|
||||
Before implementing new functionality, open a feature request describing the problem, why it belongs in OpenCode, and your proposed approach if you have one. Wait for design approval before opening the implementation PR.
|
||||
# Production
|
||||
opencode --help # Show all available commands
|
||||
opencode serve # Start headless API server
|
||||
opencode web # Start server + open web interface
|
||||
opencode <directory> # Start TUI in specific directory
|
||||
```
|
||||
|
||||
Base branches on `v2`, not `dev`, and complete the provided pull request template.
|
||||
### Running the API Server
|
||||
|
||||
### Keep It Focused
|
||||
To start the OpenCode headless API server:
|
||||
|
||||
- Keep PRs small and focused.
|
||||
- Explain the problem and why the change fixes it.
|
||||
- Check whether the functionality already exists.
|
||||
- For UI changes, include before-and-after screenshots or video.
|
||||
- For logic changes, explain what you tested and how a reviewer can verify it.
|
||||
```bash
|
||||
bun dev serve
|
||||
```
|
||||
|
||||
### Keep It Brief
|
||||
This starts the headless server on port 4096 by default. You can specify a different port:
|
||||
|
||||
Long, AI-generated PR descriptions and issues may be ignored. Write a short explanation in your own words. If the change cannot be explained briefly, the PR may be too large.
|
||||
```bash
|
||||
bun dev serve --port 8080
|
||||
```
|
||||
|
||||
### Use Conventional Titles
|
||||
### Running the Web App
|
||||
|
||||
Use `type(scope): summary`. Supported types are `feat`, `fix`, `docs`, `chore`, `refactor`, and `test`. The scope is optional.
|
||||
To test UI changes during development:
|
||||
|
||||
1. **First, start the OpenCode server** (see [Running the API Server](#running-the-api-server) section above)
|
||||
2. **Then run the web app:**
|
||||
|
||||
```bash
|
||||
bun run --cwd packages/app dev
|
||||
```
|
||||
|
||||
This starts a local dev server at http://localhost:5173 (or similar port shown in output). Most UI changes can be tested here, but the server must be running for full functionality.
|
||||
|
||||
### Running the Desktop App
|
||||
|
||||
The desktop app is an Electron application that wraps the web UI.
|
||||
|
||||
To run the desktop app in development:
|
||||
|
||||
```bash
|
||||
bun run --cwd packages/desktop dev
|
||||
```
|
||||
|
||||
To create a production build and package the app:
|
||||
|
||||
```bash
|
||||
bun run --cwd packages/desktop build
|
||||
bun run --cwd packages/desktop package
|
||||
```
|
||||
|
||||
> [!NOTE]
|
||||
> If you make changes to the API or SDK (e.g. `packages/opencode/src/server/server.ts`), run `./script/generate.ts` to regenerate the SDK and related files.
|
||||
|
||||
Please try to follow the [style guide](./AGENTS.md)
|
||||
|
||||
### Setting up a Debugger
|
||||
|
||||
Bun debugging is currently rough around the edges. We hope this guide helps you get set up and avoid some pain points.
|
||||
|
||||
The most reliable way to debug OpenCode is to run it manually in a terminal via `bun run --inspect=<url> dev ...` and attach
|
||||
your debugger via that URL. Other methods can result in breakpoints being mapped incorrectly, at least in VSCode (YMMV).
|
||||
|
||||
Caveats:
|
||||
|
||||
- If you want to run the OpenCode TUI and have breakpoints triggered in the server code, you might need to run `bun dev spawn` instead of
|
||||
the usual `bun dev`. This is because `bun dev` runs the server in a worker thread and breakpoints might not work there.
|
||||
- If `spawn` does not work for you, you can debug the server separately:
|
||||
- Debug server: `bun run --inspect=ws://localhost:6499/ --cwd packages/opencode ./src/index.ts serve --port 4096`,
|
||||
then attach TUI with `opencode attach http://localhost:4096`
|
||||
- Debug TUI: `bun run --inspect=ws://localhost:6499/ --cwd packages/opencode --conditions=browser ./src/index.ts`
|
||||
|
||||
Other tips and tricks:
|
||||
|
||||
- You might want to use `--inspect-wait` or `--inspect-brk` instead of `--inspect`, depending on your workflow
|
||||
- Specifying `--inspect=ws://localhost:6499/` on every invocation can be tiresome, you may want to `export BUN_OPTIONS=--inspect=ws://localhost:6499/` instead
|
||||
|
||||
#### VSCode Setup
|
||||
|
||||
If you use VSCode, you can use our example configurations [.vscode/settings.example.json](.vscode/settings.example.json) and [.vscode/launch.example.json](.vscode/launch.example.json).
|
||||
|
||||
Some debug methods that can be problematic:
|
||||
|
||||
- Debug configurations with `"request": "launch"` can have breakpoints incorrectly mapped and thus unusable
|
||||
- The same problem arises when running OpenCode in the VSCode `JavaScript Debug Terminal`
|
||||
|
||||
With that said, you may want to try these methods, as they might work for you.
|
||||
|
||||
## Pull Request Expectations
|
||||
|
||||
### Issue First Policy
|
||||
|
||||
**All PRs must reference an existing issue.** Before opening a PR, open an issue describing the bug or feature. This helps maintainers triage and prevents duplicate work. PRs without a linked issue may be closed without review.
|
||||
|
||||
- Use `Fixes #123` or `Closes #123` in your PR description to link the issue
|
||||
- For small fixes, a brief issue is fine - just enough context for maintainers to understand the problem
|
||||
|
||||
### General Requirements
|
||||
|
||||
- Keep pull requests small and focused
|
||||
- Explain the issue and why your change fixes it
|
||||
- Before adding new functionality, ensure it doesn't already exist elsewhere in the codebase
|
||||
|
||||
### UI Changes
|
||||
|
||||
If your PR includes UI changes, please include screenshots or videos showing the before and after. This helps maintainers review faster and gives you quicker feedback.
|
||||
|
||||
### Logic Changes
|
||||
|
||||
For non-UI changes (bug fixes, new features, refactors), explain **how you verified it works**:
|
||||
|
||||
- What did you test?
|
||||
- How can a reviewer reproduce/confirm the fix?
|
||||
|
||||
### No AI-Generated Walls of Text
|
||||
|
||||
Long, AI-generated PR descriptions and issues are not acceptable and may be ignored. Respect the maintainers' time:
|
||||
|
||||
- Write short, focused descriptions
|
||||
- Explain what changed and why in your own words
|
||||
- If you can't explain it briefly, your PR might be too large
|
||||
|
||||
### PR Titles
|
||||
|
||||
PR titles should follow conventional commit standards:
|
||||
|
||||
- `feat:` new feature or functionality
|
||||
- `fix:` bug fix
|
||||
- `docs:` documentation or README changes
|
||||
- `chore:` maintenance tasks, dependency updates, etc.
|
||||
- `refactor:` code refactoring without changing behavior
|
||||
- `test:` adding or updating tests
|
||||
|
||||
You can optionally include a scope to indicate which package is affected:
|
||||
|
||||
- `feat(app):` feature in the app package
|
||||
- `fix(desktop):` bug fix in the desktop package
|
||||
- `chore(opencode):` maintenance in the opencode package
|
||||
|
||||
Examples:
|
||||
|
||||
- `docs: update contributing guide`
|
||||
- `fix(tui): restore scroll position`
|
||||
- `feat(app): add workspace search`
|
||||
- `docs: update contributing guidelines`
|
||||
- `fix: resolve crash on startup`
|
||||
- `feat: add dark mode support`
|
||||
- `feat(app): add dark mode support`
|
||||
- `fix(desktop): resolve crash on startup`
|
||||
- `chore: bump dependency versions`
|
||||
|
||||
## Issues
|
||||
### Style Preferences
|
||||
|
||||
Bug reports and feature requests must use their issue templates. Blank issues are not allowed; ask support and how-to questions in the [Discord community](https://discord.gg/opencode).
|
||||
These are not strictly enforced, they are just general guidelines:
|
||||
|
||||
Automated checks flag missing templates, placeholder text, AI-generated walls of text, and missing meaningful content. You have two hours to correct a flagged issue before it closes automatically. Ask a maintainer if an issue was flagged incorrectly.
|
||||
- **Functions:** Keep logic within a single function unless breaking it out adds clear reuse or composition benefits.
|
||||
- **Destructuring:** Do not do unnecessary destructuring of variables.
|
||||
- **Control flow:** Avoid `else` statements.
|
||||
- **Error handling:** Prefer `.catch(...)` instead of `try`/`catch` when possible.
|
||||
- **Types:** Reach for precise types and avoid `any`.
|
||||
- **Variables:** Stick to immutable patterns and avoid `let`.
|
||||
- **Naming:** Choose concise single-word identifiers when they remain descriptive.
|
||||
- **Runtime APIs:** Use Bun helpers such as `Bun.file()` when they fit the use case.
|
||||
|
||||
## Feature Requests
|
||||
|
||||
For net-new functionality, start with a design conversation. Open an issue describing the problem, your proposed approach (optional), and why it belongs in OpenCode. The core team will help decide whether it should move forward; please wait for that approval instead of opening a feature PR directly.
|
||||
|
||||
## Trust & Vouch System
|
||||
|
||||
This project uses [vouch](https://github.com/mitchellh/vouch) to manage contributor trust. The vouch list is maintained in [`.github/VOUCHED.td`](.github/VOUCHED.td).
|
||||
|
||||
### How it works
|
||||
|
||||
- **Vouched users** are explicitly trusted contributors.
|
||||
- **Denounced users** are explicitly blocked. Issues and pull requests from denounced users are automatically closed. If you have been denounced, you can request to be unvouched by reaching out to a maintainer on [Discord](https://opencode.ai/discord)
|
||||
- **Everyone else** can participate normally — you don't need to be vouched to open issues or PRs.
|
||||
|
||||
### For maintainers
|
||||
|
||||
Collaborators with write access can manage the vouch list by commenting on any issue:
|
||||
|
||||
- `vouch` — vouch for the issue author
|
||||
- `vouch @username` — vouch for a specific user
|
||||
- `denounce` — denounce the issue author
|
||||
- `denounce @username` — denounce a specific user
|
||||
- `denounce @username <reason>` — denounce with a reason
|
||||
- `unvouch` / `unvouch @username` — remove someone from the list
|
||||
|
||||
Changes are committed automatically to `.github/VOUCHED.td`.
|
||||
|
||||
### Denouncement policy
|
||||
|
||||
Denouncement is reserved for users who repeatedly submit low-quality AI-generated contributions, spam, or otherwise act in bad faith. It is not used for disagreements or honest mistakes.
|
||||
|
||||
## Issue Requirements
|
||||
|
||||
All issues **must** use one of our issue templates:
|
||||
|
||||
- **Bug report** — for reporting bugs (requires a description)
|
||||
- **Feature request** — for suggesting enhancements (requires verification checkbox and description)
|
||||
- **Question** — for asking questions (requires the question)
|
||||
|
||||
Blank issues are not allowed. When a new issue is opened, an automated check verifies that it follows a template and meets our contributing guidelines. If an issue doesn't meet the requirements, you'll receive a comment explaining what needs to be fixed and have **2 hours** to edit the issue. After that, it will be automatically closed.
|
||||
|
||||
Issues may be flagged for:
|
||||
|
||||
- Not using a template
|
||||
- Required fields left empty or filled with placeholder text
|
||||
- AI-generated walls of text
|
||||
- Missing meaningful content
|
||||
|
||||
If you believe your issue was incorrectly flagged, let a maintainer know.
|
||||
|
||||
Binary file not shown.
Binary file not shown.
|
Before Width: | Height: | Size: 62 KiB |
+1
-1
@@ -2,7 +2,7 @@
|
||||
exact = true
|
||||
# Only install newly resolved package versions published at least 3 days ago.
|
||||
minimumReleaseAge = 259200
|
||||
minimumReleaseAgeExcludes = ["@ai-sdk/amazon-bedrock", "@ai-sdk/anthropic", "@opencode-ai/sdk", "@opentui/core", "@opentui/core-darwin-arm64", "@opentui/core-darwin-x64", "@opentui/core-linux-arm64", "@opentui/core-linux-arm64-musl", "@opentui/core-linux-x64", "@opentui/core-linux-x64-musl", "@opentui/core-win32-arm64", "@opentui/core-win32-x64", "@opentui/keymap", "@opentui/solid", "opentui-spinner", "gitlab-ai-provider", "opencode-gitlab-auth", "@ff-labs/fff-node", "@ff-labs/fff-bun", "@ff-labs/fff-bin-darwin-arm64", "@ff-labs/fff-bin-darwin-x64", "@ff-labs/fff-bin-linux-arm64-gnu", "@ff-labs/fff-bin-linux-arm64-musl", "@ff-labs/fff-bin-linux-x64-gnu", "@ff-labs/fff-bin-linux-x64-musl", "@ff-labs/fff-bin-win32-arm64", "@ff-labs/fff-bin-win32-x64", "@pierre/diffs", "@pierre/theming", "app-builder-lib", "dmg-builder", "electron-builder", "electron-publish"]
|
||||
minimumReleaseAgeExcludes = ["@ai-sdk/amazon-bedrock", "@ai-sdk/anthropic", "@opentui/core", "@opentui/core-darwin-arm64", "@opentui/core-darwin-x64", "@opentui/core-linux-arm64", "@opentui/core-linux-arm64-musl", "@opentui/core-linux-x64", "@opentui/core-linux-x64-musl", "@opentui/core-win32-arm64", "@opentui/core-win32-x64", "@opentui/keymap", "@opentui/solid", "opentui-spinner", "gitlab-ai-provider", "opencode-gitlab-auth", "@ff-labs/fff-node", "@ff-labs/fff-bun", "@ff-labs/fff-bin-darwin-arm64", "@ff-labs/fff-bin-darwin-x64", "@ff-labs/fff-bin-linux-arm64-gnu", "@ff-labs/fff-bin-linux-arm64-musl", "@ff-labs/fff-bin-linux-x64-gnu", "@ff-labs/fff-bin-linux-x64-musl", "@ff-labs/fff-bin-win32-arm64", "@ff-labs/fff-bin-win32-x64", "@pierre/diffs", "@pierre/theming", "app-builder-lib", "dmg-builder", "electron-builder", "electron-publish"]
|
||||
|
||||
[test]
|
||||
root = "./do-not-run-tests-from-root"
|
||||
|
||||
@@ -495,6 +495,7 @@ async function subscribeSessionEvents() {
|
||||
console.log("Subscribing to session events...")
|
||||
|
||||
const TOOL: Record<string, [string, string]> = {
|
||||
todowrite: ["Todo", "\x1b[33m\x1b[1m"],
|
||||
bash: ["Bash", "\x1b[31m\x1b[1m"],
|
||||
edit: ["Edit", "\x1b[32m\x1b[1m"],
|
||||
glob: ["Glob", "\x1b[34m\x1b[1m"],
|
||||
|
||||
+1
-1
@@ -15,6 +15,6 @@
|
||||
"@actions/github": "6.0.1",
|
||||
"@octokit/graphql": "9.0.1",
|
||||
"@octokit/rest": "catalog:",
|
||||
"@opencode-ai/sdk": "1.18.5"
|
||||
"@opencode-ai/sdk": "workspace:*"
|
||||
}
|
||||
}
|
||||
|
||||
+2
-6
@@ -288,12 +288,8 @@ new sst.cloudflare.x.SolidStart("Console", {
|
||||
server: {
|
||||
placement: { region: "aws:us-east-2" },
|
||||
transform: {
|
||||
worker: (args) => {
|
||||
args.compatibilityFlags = $resolve(args.compatibilityFlags).apply((flags) => [
|
||||
...(flags ?? []),
|
||||
"global_fetch_strictly_public",
|
||||
])
|
||||
args.tailConsumers = [{ service: logProcessor.nodes.worker.scriptName }]
|
||||
worker: {
|
||||
tailConsumers: [{ service: logProcessor.nodes.worker.scriptName }],
|
||||
},
|
||||
},
|
||||
},
|
||||
|
||||
@@ -67,10 +67,6 @@ const athenaWorkgroup = new aws.athena.Workgroup("LakeAthenaWorkgroup", {
|
||||
configuration: {
|
||||
enforceWorkgroupConfiguration: true,
|
||||
publishCloudwatchMetricsEnabled: true,
|
||||
// Athena bills $5/TB scanned; kill any query that would scan more than 2 TB
|
||||
// so a regression cannot silently burn money. Stats sync full passes scan
|
||||
// ~250 GB as of 2026-07.
|
||||
bytesScannedCutoffPerQuery: 2 * 1024 ** 4,
|
||||
resultConfiguration: {
|
||||
outputLocation: $interpolate`s3://${athenaResultsBucket.bucket}/`,
|
||||
},
|
||||
|
||||
@@ -8,25 +8,6 @@ export const zoneID = "430ba34c138cfb5360826c4909f99be8"
|
||||
export const awsStage = $app.stage === "production" ? "production" : "dev"
|
||||
export const deployAws = $app.stage === awsStage
|
||||
|
||||
if ($app.stage === "production") {
|
||||
new cloudflare.DnsRecord("TrustCenter", {
|
||||
zoneId: zoneID,
|
||||
name: "trust.opencode.ai",
|
||||
type: "CNAME",
|
||||
content: "3a69a5bb27875189.vercel-dns-016.com",
|
||||
proxied: false,
|
||||
ttl: 60,
|
||||
})
|
||||
|
||||
new cloudflare.DnsRecord("TrustCenterVerification", {
|
||||
zoneId: zoneID,
|
||||
name: "opencode.ai",
|
||||
type: "TXT",
|
||||
content: "compai-domain-verification=org_6993a99c6200a2d642bb115d",
|
||||
ttl: 60,
|
||||
})
|
||||
}
|
||||
|
||||
new cloudflare.RegionalHostname("RegionalHostname", {
|
||||
hostname: domain,
|
||||
regionKey: "us",
|
||||
|
||||
+1
-3
@@ -185,9 +185,7 @@ export const statSync = new sst.aws.Service("StatsSyncService", {
|
||||
cluster: lakeCluster,
|
||||
architecture: "arm64",
|
||||
cpu: "0.25 vCPU",
|
||||
// 0.5 GB caused an OOM crash loop: every restart immediately re-ran the 4 Athena
|
||||
// stats queries (~$5/pass) every ~5 minutes instead of hourly.
|
||||
memory: "2 GB",
|
||||
memory: "0.5 GB",
|
||||
image: {
|
||||
context: ".",
|
||||
dockerfile: "packages/stats/server/Dockerfile",
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
#!/usr/bin/env bash
|
||||
set -euo pipefail
|
||||
APP=opencode2
|
||||
APP=opencode
|
||||
|
||||
MUTED='\033[0;2m'
|
||||
RED='\033[0;31m'
|
||||
@@ -15,14 +15,14 @@ Usage: install.sh [options]
|
||||
|
||||
Options:
|
||||
-h, --help Display this help message
|
||||
-v, --version <version> Install a specific version (e.g., 0.0.0-next-17236)
|
||||
-v, --version <version> Install a specific version (e.g., 1.0.180)
|
||||
-b, --binary <path> Install from a local binary instead of downloading
|
||||
--no-modify-path Don't modify shell config files (.zshrc, .bashrc, etc.)
|
||||
|
||||
Examples:
|
||||
curl -fsSL https://raw.githubusercontent.com/anomalyco/opencode/v2/install | bash
|
||||
curl -fsSL https://raw.githubusercontent.com/anomalyco/opencode/v2/install | bash -s -- --version 0.0.0-next-17236
|
||||
./install --binary /path/to/opencode2
|
||||
curl -fsSL https://opencode.ai/install | bash
|
||||
curl -fsSL https://opencode.ai/install | bash -s -- --version 1.0.180
|
||||
./install --binary /path/to/opencode
|
||||
EOF
|
||||
}
|
||||
|
||||
@@ -109,6 +109,11 @@ else
|
||||
;;
|
||||
esac
|
||||
|
||||
archive_ext=".zip"
|
||||
if [ "$os" = "linux" ]; then
|
||||
archive_ext=".tar.gz"
|
||||
fi
|
||||
|
||||
is_musl=false
|
||||
if [ "$os" = "linux" ]; then
|
||||
if [ -f /etc/alpine-release ]; then
|
||||
@@ -160,42 +165,42 @@ else
|
||||
target="$target-musl"
|
||||
fi
|
||||
|
||||
if ! command -v tar >/dev/null 2>&1; then
|
||||
echo -e "${RED}Error: 'tar' is required but not installed.${NC}"
|
||||
exit 1
|
||||
filename="$APP-$target$archive_ext"
|
||||
|
||||
|
||||
if [ "$os" = "linux" ]; then
|
||||
if ! command -v tar >/dev/null 2>&1; then
|
||||
echo -e "${RED}Error: 'tar' is required but not installed.${NC}"
|
||||
exit 1
|
||||
fi
|
||||
else
|
||||
if ! command -v unzip >/dev/null 2>&1; then
|
||||
echo -e "${RED}Error: 'unzip' is required but not installed.${NC}"
|
||||
exit 1
|
||||
fi
|
||||
fi
|
||||
|
||||
if [ -z "$requested_version" ]; then
|
||||
metadata=$(curl -fsSL https://registry.npmjs.org/@opencode-ai%2fcli/next || true)
|
||||
specific_version=$(echo "$metadata" | sed -n 's/.*"version":"\([^"]*\)".*/\1/p')
|
||||
url="https://github.com/anomalyco/opencode/releases/latest/download/$filename"
|
||||
specific_version=$(curl -s https://api.github.com/repos/anomalyco/opencode/releases/latest | sed -n 's/.*"tag_name": *"v\([^"]*\)".*/\1/p')
|
||||
|
||||
if [ -z "$specific_version" ]; then
|
||||
if [[ $? -ne 0 || -z "$specific_version" ]]; then
|
||||
echo -e "${RED}Failed to fetch version information${NC}"
|
||||
exit 1
|
||||
fi
|
||||
else
|
||||
# Strip leading 'v' if present
|
||||
requested_version="${requested_version#v}"
|
||||
url="https://github.com/anomalyco/opencode/releases/download/v${requested_version}/$filename"
|
||||
specific_version=$requested_version
|
||||
fi
|
||||
|
||||
package_name="@opencode-ai/cli-$target"
|
||||
http_status=$(curl -s -o /dev/null -w "%{http_code}" "https://registry.npmjs.org/@opencode-ai%2fcli-$target/$specific_version" || true)
|
||||
if [ "$http_status" = "404" ]; then
|
||||
echo -e "${RED}Error: Version ${specific_version} is not available for $target${NC}"
|
||||
echo -e "${MUTED}Available versions: https://www.npmjs.com/package/$package_name?activeTab=versions${NC}"
|
||||
exit 1
|
||||
fi
|
||||
if [ "$http_status" != "200" ]; then
|
||||
echo -e "${RED}Failed to fetch package information${NC}"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
filename="cli-$target-$specific_version.tgz"
|
||||
url="https://registry.npmjs.org/$package_name/-/$filename"
|
||||
binary_name="$APP"
|
||||
if [ "$os" = "windows" ]; then
|
||||
binary_name="$APP.exe"
|
||||
# Verify the release exists before downloading
|
||||
http_status=$(curl -sI -o /dev/null -w "%{http_code}" "https://github.com/anomalyco/opencode/releases/tag/v${requested_version}")
|
||||
if [ "$http_status" = "404" ]; then
|
||||
echo -e "${RED}Error: Release v${requested_version} not found${NC}"
|
||||
echo -e "${MUTED}Available releases: https://github.com/anomalyco/opencode/releases${NC}"
|
||||
exit 1
|
||||
fi
|
||||
fi
|
||||
fi
|
||||
|
||||
@@ -214,13 +219,11 @@ print_message() {
|
||||
}
|
||||
|
||||
check_version() {
|
||||
if command -v "$APP" >/dev/null 2>&1; then
|
||||
opencode_path=$(which "$APP")
|
||||
if command -v opencode >/dev/null 2>&1; then
|
||||
opencode_path=$(which opencode)
|
||||
|
||||
## Check the installed version
|
||||
installed_version=$("$APP" --version 2>/dev/null || echo "")
|
||||
installed_version="${installed_version##* }"
|
||||
installed_version="${installed_version#v}"
|
||||
installed_version=$(opencode --version 2>/dev/null || echo "")
|
||||
|
||||
if [[ "$installed_version" != "$specific_version" ]]; then
|
||||
print_message info "${MUTED}Installed version: ${NC}$installed_version."
|
||||
@@ -284,7 +287,7 @@ download_with_progress() {
|
||||
trap "trap - RETURN; rm -f \"$tracefile\"; printf '\033[?25h' >&4; exec 4>&-" RETURN
|
||||
|
||||
(
|
||||
curl --trace-ascii "$tracefile" -fsL -o "$output" "$url"
|
||||
curl --trace-ascii "$tracefile" -s -L -o "$output" "$url"
|
||||
) &
|
||||
local curl_pid=$!
|
||||
|
||||
@@ -322,25 +325,30 @@ download_with_progress() {
|
||||
}
|
||||
|
||||
download_and_install() {
|
||||
print_message info "\n${MUTED}Installing ${NC}$APP ${MUTED}version: ${NC}$specific_version"
|
||||
print_message info "\n${MUTED}Installing ${NC}opencode ${MUTED}version: ${NC}$specific_version"
|
||||
local tmp_dir="${TMPDIR:-/tmp}/opencode_install_$$"
|
||||
mkdir -p "$tmp_dir"
|
||||
|
||||
if [[ "$os" == "windows" ]] || ! [ -t 2 ] || ! download_with_progress "$url" "$tmp_dir/$filename"; then
|
||||
# Fallback to standard curl on Windows, non-TTY environments, or if custom progress fails
|
||||
curl -f -# -L -o "$tmp_dir/$filename" "$url"
|
||||
curl -# -L -o "$tmp_dir/$filename" "$url"
|
||||
fi
|
||||
|
||||
tar -xzf "$tmp_dir/$filename" -C "$tmp_dir"
|
||||
mv "$tmp_dir/package/bin/$binary_name" "$INSTALL_DIR"
|
||||
chmod 755 "${INSTALL_DIR}/$binary_name"
|
||||
if [ "$os" = "linux" ]; then
|
||||
tar -xzf "$tmp_dir/$filename" -C "$tmp_dir"
|
||||
else
|
||||
unzip -q "$tmp_dir/$filename" -d "$tmp_dir"
|
||||
fi
|
||||
|
||||
mv "$tmp_dir/opencode" "$INSTALL_DIR"
|
||||
chmod 755 "${INSTALL_DIR}/opencode"
|
||||
rm -rf "$tmp_dir"
|
||||
}
|
||||
|
||||
install_from_binary() {
|
||||
print_message info "\n${MUTED}Installing ${NC}$APP ${MUTED}from: ${NC}$binary_path"
|
||||
cp "$binary_path" "${INSTALL_DIR}/$APP"
|
||||
chmod 755 "${INSTALL_DIR}/$APP"
|
||||
print_message info "\n${MUTED}Installing ${NC}opencode ${MUTED}from: ${NC}$binary_path"
|
||||
cp "$binary_path" "${INSTALL_DIR}/opencode"
|
||||
chmod 755 "${INSTALL_DIR}/opencode"
|
||||
}
|
||||
|
||||
if [ -n "$binary_path" ]; then
|
||||
@@ -445,8 +453,8 @@ echo -e ""
|
||||
echo -e "${MUTED}OpenCode includes free models, to start:${NC}"
|
||||
echo -e ""
|
||||
echo -e "cd <project> ${MUTED}# Open directory${NC}"
|
||||
echo -e "opencode2 ${MUTED}# Run command${NC}"
|
||||
echo -e "opencode ${MUTED}# Run command${NC}"
|
||||
echo -e ""
|
||||
echo -e "${MUTED}For more information visit ${NC}https://opencode.ai/v2/docs"
|
||||
echo -e "${MUTED}For more information visit ${NC}https://opencode.ai/docs"
|
||||
echo -e ""
|
||||
echo -e ""
|
||||
|
||||
+33
-66
@@ -8,8 +8,6 @@
|
||||
makeWrapper,
|
||||
writableTmpDirAsHomeHook,
|
||||
autoPatchelfHook,
|
||||
copyDesktopItems,
|
||||
makeDesktopItem,
|
||||
opencode,
|
||||
}:
|
||||
let
|
||||
@@ -29,12 +27,9 @@ stdenv.mkDerivation (finalAttrs: {
|
||||
nodejs
|
||||
makeWrapper
|
||||
writableTmpDirAsHomeHook
|
||||
]
|
||||
++ lib.optionals stdenv.hostPlatform.isLinux [
|
||||
] ++ lib.optionals stdenv.hostPlatform.isLinux [
|
||||
autoPatchelfHook
|
||||
copyDesktopItems
|
||||
]
|
||||
++ lib.optionals stdenv.hostPlatform.isDarwin [
|
||||
] ++ lib.optionals stdenv.hostPlatform.isDarwin [
|
||||
# Ad-hoc sign the .app: --config.mac.identity=null below skips signing.
|
||||
darwin.autoSignDarwinBinariesHook
|
||||
];
|
||||
@@ -43,37 +38,20 @@ stdenv.mkDerivation (finalAttrs: {
|
||||
(lib.getLib stdenv.cc.cc)
|
||||
];
|
||||
|
||||
desktopItems = lib.optional stdenv.hostPlatform.isLinux (makeDesktopItem {
|
||||
name = "ai.opencode.desktop";
|
||||
desktopName = "OpenCode";
|
||||
exec = "opencode-desktop %U";
|
||||
icon = "ai.opencode.desktop";
|
||||
# Electron 41 derives X11 WM_CLASS from app.name.
|
||||
startupWMClass = "OpenCode";
|
||||
categories = [ "Development" ];
|
||||
});
|
||||
|
||||
env = opencode.env // {
|
||||
ELECTRON_SKIP_BINARY_DOWNLOAD = "1";
|
||||
};
|
||||
|
||||
postPatch =
|
||||
# NOTE: Relax Bun version check to be a warning instead of an error
|
||||
''
|
||||
substituteInPlace packages/script/src/index.ts \
|
||||
--replace-fail 'throw new Error(`This script requires bun@''${expectedBunVersionRange}' \
|
||||
'console.warn(`Warning: This script requires bun@''${expectedBunVersionRange}'
|
||||
''
|
||||
# https://github.com/electron/electron/issues/31121
|
||||
# mac builds use a .app bundle which doesnt have this issue
|
||||
+ lib.optionalString stdenv.isLinux ''
|
||||
BASE_PATH=packages/desktop
|
||||
FILES=(src/main/windows.ts)
|
||||
for file in "''${FILES[@]}"; do
|
||||
substituteInPlace $BASE_PATH/$file \
|
||||
--replace-fail "process.resourcesPath" "'$out/opt/opencode-desktop/resources'"
|
||||
done
|
||||
'';
|
||||
# https://github.com/electron/electron/issues/31121
|
||||
# mac builds use a .app bundle which doesnt have this issue
|
||||
postPatch = lib.optionalString stdenv.isLinux ''
|
||||
BASE_PATH=packages/desktop
|
||||
FILES=(src/main/windows.ts)
|
||||
for file in "''${FILES[@]}"; do
|
||||
substituteInPlace $BASE_PATH/$file \
|
||||
--replace-fail "process.resourcesPath" "'$out/opt/opencode-desktop/resources'"
|
||||
done
|
||||
'';
|
||||
|
||||
preBuild = ''
|
||||
cp -r "${electron.dist}" $HOME/.electron-dist
|
||||
@@ -98,38 +76,27 @@ stdenv.mkDerivation (finalAttrs: {
|
||||
runHook postBuild
|
||||
'';
|
||||
|
||||
installPhase = ''
|
||||
runHook preInstall
|
||||
''
|
||||
+ lib.optionalString stdenv.hostPlatform.isDarwin ''
|
||||
mkdir -p $out/Applications
|
||||
mv dist/mac*/*.app $out/Applications
|
||||
makeWrapper "$out/Applications/OpenCode.app/Contents/MacOS/OpenCode" $out/bin/opencode-desktop
|
||||
''
|
||||
+ lib.optionalString stdenv.hostPlatform.isLinux ''
|
||||
mkdir -p $out/opt/opencode-desktop
|
||||
cp -r dist/linux*-unpacked/{resources,LICENSE*} $out/opt/opencode-desktop
|
||||
install -Dm644 resources/icons/32x32.png \
|
||||
"$out/share/icons/hicolor/32x32/apps/ai.opencode.desktop.png"
|
||||
install -Dm644 resources/icons/64x64.png \
|
||||
"$out/share/icons/hicolor/64x64/apps/ai.opencode.desktop.png"
|
||||
install -Dm644 resources/icons/128x128.png \
|
||||
"$out/share/icons/hicolor/128x128/apps/ai.opencode.desktop.png"
|
||||
install -Dm644 resources/icons/128x128@2x.png \
|
||||
"$out/share/icons/hicolor/256x256/apps/ai.opencode.desktop.png"
|
||||
install -Dm644 resources/icons/icon.png \
|
||||
"$out/share/icons/hicolor/512x512/apps/ai.opencode.desktop.png"
|
||||
install -Dm644 resources/ai.opencode.desktop.metainfo.xml \
|
||||
"$out/share/metainfo/ai.opencode.desktop.metainfo.xml"
|
||||
makeWrapper ${lib.getExe electron} $out/bin/opencode-desktop \
|
||||
--inherit-argv0 \
|
||||
--set ELECTRON_FORCE_IS_PACKAGED 1 \
|
||||
--add-flags $out/opt/opencode-desktop/resources/app.asar \
|
||||
--add-flags "\''${NIXOS_OZONE_WL:+\''${WAYLAND_DISPLAY:+--ozone-platform-hint=auto --enable-features=WaylandWindowDecorations --enable-wayland-ime=true}}"
|
||||
''
|
||||
+ ''
|
||||
runHook postInstall
|
||||
'';
|
||||
installPhase =
|
||||
''
|
||||
runHook preInstall
|
||||
''
|
||||
+ lib.optionalString stdenv.hostPlatform.isDarwin ''
|
||||
mkdir -p $out/Applications
|
||||
mv dist/mac*/*.app $out/Applications
|
||||
makeWrapper "$out/Applications/OpenCode.app/Contents/MacOS/OpenCode" $out/bin/opencode-desktop
|
||||
''
|
||||
+ lib.optionalString stdenv.hostPlatform.isLinux ''
|
||||
mkdir -p $out/opt/opencode-desktop
|
||||
cp -r dist/linux*-unpacked/{resources,LICENSE*} $out/opt/opencode-desktop
|
||||
makeWrapper ${lib.getExe electron} $out/bin/opencode-desktop \
|
||||
--inherit-argv0 \
|
||||
--set ELECTRON_FORCE_IS_PACKAGED 1 \
|
||||
--add-flags $out/opt/opencode-desktop/resources/app.asar \
|
||||
--add-flags "\''${NIXOS_OZONE_WL:+\''${WAYLAND_DISPLAY:+--ozone-platform-hint=auto --enable-features=WaylandWindowDecorations --enable-wayland-ime=true}}"
|
||||
''
|
||||
+ ''
|
||||
runHook postInstall
|
||||
'';
|
||||
|
||||
autoPatchelfIgnoreMissingDeps = [
|
||||
"libc.musl-x86_64.so.1"
|
||||
|
||||
+4
-4
@@ -1,8 +1,8 @@
|
||||
{
|
||||
"nodeModules": {
|
||||
"x86_64-linux": "sha256-uduwrM143NDSc+tXsi4lVVfoMll2a3BDHRUjuO7GB68=",
|
||||
"aarch64-linux": "sha256-6DUda78XdXY6DP86lIUkweSjys3iG4Y4mo1PiaNuXbg=",
|
||||
"aarch64-darwin": "sha256-AkJwfLULLZVwwz+XU1QcFUZoIS7oVPCn+n/MXEaxrqE=",
|
||||
"x86_64-darwin": "sha256-hAxKGdiITTxQ2uujQt6prNjo3NxGAMMeo+9HlMWK6GU="
|
||||
"x86_64-linux": "sha256-KTyd2ISQ4n1E3tm1LMEHtz+rKRZga+SONn5J6H0pheQ=",
|
||||
"aarch64-linux": "sha256-ZIeaaqRr4JorEmmwFYuitxfYEAtrP8C6IlD+pmFRko0=",
|
||||
"aarch64-darwin": "sha256-Po8FISoBzUMwJSn6nw243/hLT/gAQCrm5HbwXe2uA0g=",
|
||||
"x86_64-darwin": "sha256-5TZzrCnvg1z00YP6O9U0SXjL04r+VmF7LVrqQ9BbSn4="
|
||||
}
|
||||
}
|
||||
|
||||
+19
-32
@@ -8,15 +8,13 @@
|
||||
"packageManager": "bun@1.3.14",
|
||||
"scripts": {
|
||||
"dev": "bun run --cwd packages/cli --conditions=browser src/index.ts",
|
||||
"dev:live": "OPENCODE_TUI_CHANNEL=next OPENCODE_PASSWORD=\"$(opencode2 service get password)\" bun run dev --server \"$(opencode2 service status)\"",
|
||||
"dev:desktop": "bun --cwd packages/desktop dev",
|
||||
"dev:web": "bun --cwd packages/app dev",
|
||||
"dev:console": "ulimit -n 10240 2>/dev/null; bun run --cwd packages/console/app dev",
|
||||
"dev:stats": "bun sst shell --stage=production -- bun run --cwd packages/stats/app dev",
|
||||
"dev:www": "bun run --cwd packages/www dev",
|
||||
"dev:storybook": "bun --cwd packages/storybook storybook",
|
||||
"lint": "oxlint",
|
||||
"lint:effect-patterns": "ast-grep scan -c script/ast-grep/sgconfig.yml packages/util/src packages/core/src packages/server/src packages/protocol/src packages/cli/src",
|
||||
"lint:effect-patterns": "ast-grep scan -c script/ast-grep/sgconfig.yml packages/core/src packages/server/src packages/protocol/src packages/cli/src",
|
||||
"test:lint-rules": "ast-grep test -c script/ast-grep/sgconfig.yml",
|
||||
"typecheck": "bun turbo typecheck --concurrency=3",
|
||||
"typecheck:profile": "bun script/profile-typecheck.ts",
|
||||
@@ -26,7 +24,6 @@
|
||||
"prepare": "husky",
|
||||
"random": "echo 'Random script'",
|
||||
"sso": "aws sso login --sso-session=opencode --no-browser",
|
||||
"translate:app": "bun run script/translate-app.ts",
|
||||
"test": "echo 'do not run tests from root' && exit 1"
|
||||
},
|
||||
"workspaces": {
|
||||
@@ -34,27 +31,26 @@
|
||||
"packages/*",
|
||||
"packages/console/*",
|
||||
"packages/stats/*",
|
||||
"packages/sdk/js",
|
||||
"packages/slack"
|
||||
],
|
||||
"catalog": {
|
||||
"@effect/opentelemetry": "4.0.0-beta.101",
|
||||
"@effect/platform-node": "4.0.0-beta.101",
|
||||
"@effect/sql-sqlite-bun": "4.0.0-beta.101",
|
||||
"@effect/opentelemetry": "4.0.0-beta.83",
|
||||
"@effect/platform-node": "4.0.0-beta.83",
|
||||
"@effect/sql-sqlite-bun": "4.0.0-beta.83",
|
||||
"@npmcli/arborist": "9.4.0",
|
||||
"@types/bun": "1.3.13",
|
||||
"@types/cross-spawn": "6.0.6",
|
||||
"@octokit/rest": "22.0.0",
|
||||
"@hono/standard-validator": "0.2.0",
|
||||
"@hono/zod-validator": "0.4.2",
|
||||
"@opentui/core": "0.5.3",
|
||||
"@opentui/keymap": "0.5.3",
|
||||
"@opentui/solid": "0.5.3",
|
||||
"@tanstack/solid-virtual": "3.13.32",
|
||||
"@opentui/core": "0.4.3",
|
||||
"@opentui/keymap": "0.4.3",
|
||||
"@opentui/solid": "0.4.3",
|
||||
"@tanstack/solid-virtual": "3.13.28",
|
||||
"@shikijs/stream": "4.2.0",
|
||||
"@standard-schema/spec": "1.1.0",
|
||||
"ulid": "3.0.1",
|
||||
"@kobalte/core": "0.13.11",
|
||||
"@corvu/drawer": "0.2.4",
|
||||
"@types/luxon": "3.7.1",
|
||||
"@types/node": "24.12.2",
|
||||
"@types/semver": "7.7.1",
|
||||
@@ -70,15 +66,14 @@
|
||||
"dompurify": "3.3.1",
|
||||
"drizzle-kit": "1.0.0-rc.2",
|
||||
"drizzle-orm": "1.0.0-rc.2",
|
||||
"effect": "4.0.0-beta.101",
|
||||
"effect": "4.0.0-beta.83",
|
||||
"ai": "6.0.168",
|
||||
"cross-spawn": "7.0.6",
|
||||
"hono": "4.10.7",
|
||||
"hono-openapi": "1.1.2",
|
||||
"fuzzysort": "3.1.0",
|
||||
"get-east-asian-width": "1.6.0",
|
||||
"luxon": "3.6.1",
|
||||
"marked": "18.0.7",
|
||||
"marked": "17.0.1",
|
||||
"marked-shiki": "1.2.1",
|
||||
"remend": "1.3.0",
|
||||
"@playwright/test": "1.59.1",
|
||||
@@ -87,11 +82,9 @@
|
||||
"@typescript/native-preview": "7.0.0-dev.20251207.1",
|
||||
"zod": "4.1.8",
|
||||
"remeda": "2.26.0",
|
||||
"resolve.exports": "2.0.3",
|
||||
"sst": "4.13.1",
|
||||
"shiki": "4.2.0",
|
||||
"solid-list": "0.3.0",
|
||||
"string-width": "7.2.0",
|
||||
"tailwindcss": "4.1.11",
|
||||
"vite": "7.1.4",
|
||||
"@solidjs/meta": "0.29.4",
|
||||
@@ -100,7 +93,6 @@
|
||||
"@sentry/solid": "10.36.0",
|
||||
"@sentry/vite-plugin": "4.6.0",
|
||||
"solid-js": "1.9.10",
|
||||
"solid-sonner": "0.3.1",
|
||||
"vite-plugin-solid": "2.11.10",
|
||||
"@lydell/node-pty": "1.2.0-beta.12"
|
||||
}
|
||||
@@ -108,8 +100,6 @@
|
||||
"devDependencies": {
|
||||
"@actions/artifact": "5.0.1",
|
||||
"@ast-grep/cli": "0.44.0",
|
||||
"@types/react": "19.2.17",
|
||||
"@types/react-dom": "19.2.3",
|
||||
"@tsconfig/bun": "catalog:",
|
||||
"@types/mime-types": "3.0.1",
|
||||
"@typescript/native-preview": "catalog:",
|
||||
@@ -120,14 +110,13 @@
|
||||
"prettier": "3.6.2",
|
||||
"semver": "^7.6.0",
|
||||
"sst": "catalog:",
|
||||
"turbo": "2.10.2",
|
||||
"vitest": "4.1.10"
|
||||
"turbo": "2.10.2"
|
||||
},
|
||||
"dependencies": {
|
||||
"@aws-sdk/client-s3": "3.933.0",
|
||||
"@opencode-ai/plugin": "workspace:*",
|
||||
"@opencode-ai/script": "workspace:*",
|
||||
"@opencode-ai/sdk": "1.18.5",
|
||||
"@opencode-ai/sdk": "workspace:*",
|
||||
"heap-snapshot-toolkit": "1.1.3",
|
||||
"typescript": "catalog:"
|
||||
},
|
||||
@@ -155,24 +144,22 @@
|
||||
"@opentui/keymap": "catalog:",
|
||||
"@opentui/solid": "catalog:",
|
||||
"@types/bun": "catalog:",
|
||||
"@types/node": "catalog:",
|
||||
"effect": "catalog:"
|
||||
"@types/node": "catalog:"
|
||||
},
|
||||
"patchedDependencies": {
|
||||
"@ai-sdk/openai-compatible@2.0.41": "patches/@ai-sdk%2Fopenai-compatible@2.0.41.patch",
|
||||
"@dnd-kit/dom@0.5.0": "patches/@dnd-kit%2Fdom@0.5.0.patch",
|
||||
"@ff-labs/fff-bun@0.9.3": "patches/@ff-labs%2Ffff-bun@0.9.3.patch",
|
||||
"@npmcli/agent@4.0.2": "patches/@npmcli%2Fagent@4.0.2.patch",
|
||||
"@silvia-odwyer/photon-node@0.3.4": "patches/@silvia-odwyer%2Fphoton-node@0.3.4.patch",
|
||||
"@standard-community/standard-openapi@0.2.9": "patches/@standard-community%2Fstandard-openapi@0.2.9.patch",
|
||||
"solid-js@1.9.10": "patches/solid-js@1.9.10.patch",
|
||||
"@ai-sdk/mistral@3.0.51": "patches/@ai-sdk%2Fmistral@3.0.51.patch",
|
||||
"@ai-sdk/xai@3.0.82": "patches/@ai-sdk%2Fxai@3.0.82.patch",
|
||||
"gcp-metadata@8.1.2": "patches/gcp-metadata@8.1.2.patch",
|
||||
"pacote@21.5.0": "patches/pacote@21.5.0.patch",
|
||||
"@ai-sdk/google@3.0.73": "patches/@ai-sdk%2Fgoogle@3.0.73.patch",
|
||||
"@tanstack/solid-virtual@3.13.28": "patches/@tanstack%2Fsolid-virtual@3.13.28.patch",
|
||||
"@pierre/trees@1.0.0-beta.4": "patches/@pierre%2Ftrees@1.0.0-beta.4.patch",
|
||||
"@modelcontextprotocol/sdk@1.29.0": "patches/@modelcontextprotocol%2Fsdk@1.29.0.patch",
|
||||
"effect@4.0.0-beta.101": "patches/effect@4.0.0-beta.101.patch",
|
||||
"@tanstack/virtual-core@3.17.3": "patches/@tanstack%2Fvirtual-core@3.17.3.patch",
|
||||
"@ff-labs/fff-bun@0.10.1": "patches/@ff-labs%2Ffff-bun@0.10.1.patch"
|
||||
"@tanstack/virtual-core@3.17.0": "patches/@tanstack%2Fvirtual-core@3.17.0.patch",
|
||||
"effect@4.0.0-beta.83": "patches/effect@4.0.0-beta.83.patch"
|
||||
}
|
||||
}
|
||||
|
||||
@@ -1,339 +0,0 @@
|
||||
# AI Package Guide
|
||||
|
||||
## Effect
|
||||
|
||||
- Prefer `HttpClient.HttpClient` / `HttpClientResponse.HttpClientResponse` over web `fetch` / `Response` at package boundaries.
|
||||
- Use `Stream.Stream` for streaming data flow. Avoid ad hoc async generators or manual web reader loops unless an Effect `Stream` API cannot model the behavior.
|
||||
- Use Effect Schema codecs for JSON encode/decode (`Schema.fromJsonString(...)`) instead of direct `JSON.parse` / `JSON.stringify` in implementation code.
|
||||
- In `Effect.gen`, yield yieldable errors directly (`return yield* new MyError(...)`) instead of `Effect.fail(new MyError(...))`.
|
||||
- Use `Effect.void` instead of `Effect.succeed(undefined)` when the successful value is intentionally void.
|
||||
|
||||
## Conventions
|
||||
|
||||
Per-type constructors live on the type, not as top-level re-exports. Use `Message.system(...)`, `Message.user(...)`, `Message.assistant(...)`, `Message.tool(...)`, `LanguageModel.make(...)`, `ToolDefinition.make(...)`, `ToolCallPart.make(...)`, `ToolResultPart.make(...)`, `ToolChoice.make(...)`, `ToolChoice.named(...)`, `SystemPart.make(...)`, and `GenerationOptions.make(...)` directly. The top-level `LLM` namespace is reserved for request-shaped call APIs: `LLM.request`, `LLM.generate`, `LLM.stream`, and `LLM.generateObject`. Use `LLMRequest.update(...)` when deriving canonical request data; do not add a duplicate `LLM.updateRequest(...)` path. Two ways to construct the same thing is one too many.
|
||||
|
||||
- Keep provider-defined string enums forward-compatible. Expose known values for autocomplete while accepting future values with `Known | (string & {})`; use `Schema.String` at runtime unless rejecting unknown values is required for correctness.
|
||||
|
||||
## Tests
|
||||
|
||||
- Use `testEffect(...)` from `test/lib/effect.ts` for tests requiring Effect layers.
|
||||
- Keep provider tests fixture-first. Live provider calls must stay behind `RECORD=true` and required API-key checks.
|
||||
|
||||
## Architecture
|
||||
|
||||
This package is an Effect Schema-first LLM core. The Schema classes in `src/schema/` are the canonical runtime data model. Convenience functions in `src/llm.ts` are thin constructors that return those same Schema class instances; they should improve callsites without creating a second model.
|
||||
|
||||
Session integration lives in `packages/core/src/session`: `runner/llm.ts` owns orchestration, `model-request.ts` lowers Session state into `LLMRequest`, and `model-transport.ts` selects transport behavior.
|
||||
|
||||
Keep this package independent of Session concerns. Session auth, permissions, plugins, telemetry headers, and runtime selection belong in Core.
|
||||
|
||||
### Request Flow
|
||||
|
||||
The intended callsite is:
|
||||
|
||||
```ts
|
||||
const request = LLM.request({
|
||||
model: OpenAI.configure({ apiKey }).responses("gpt-4o-mini"),
|
||||
system: "You are concise.",
|
||||
prompt: "Say hello.",
|
||||
})
|
||||
|
||||
const response = yield * LLMClient.generate(request)
|
||||
```
|
||||
|
||||
`LLM.request(...)` builds an `LLMRequest`. `LLMClient.generate(...)` reads the executable route carried by `request.model.route`, builds the provider-native body, asks the route's transport for a real `HttpClientRequest.HttpClientRequest`, sends it through `RequestExecutor.Service`, parses the provider stream into common `LLMEvent`s, and finally returns an `LLMResponse`.
|
||||
|
||||
Use `LLMClient.stream(request)` when callers want incremental `LLMEvent`s. Use `LLMClient.generate(request)` when callers want those same events collected into an `LLMResponse`.
|
||||
|
||||
Filter or narrow `LLMEvent` streams with `LLMEvent.is.*` (camelCase guards, e.g. `events.filter(LLMEvent.is.toolCall)`). The kebab-case `LLMEvent.guards["tool-call"]` form also works but prefer `is.*` in new code.
|
||||
|
||||
### Routes
|
||||
|
||||
A route is the registered, runnable composition of four orthogonal pieces:
|
||||
|
||||
- **`Protocol`** (`src/route/protocol.ts`) — semantic API contract. Owns request body construction (`body.from`), the body schema (`body.schema`), the streaming-event schema (`stream.event`), and the event-to-`LLMEvent` state machine (`stream.step`). `Route.make(...)` validates and JSON-encodes the body from `body.schema` and decodes frames with `stream.event`. Examples: `OpenAIChat.protocol`, `OpenResponses.protocol`, `OpenAIResponses.protocol`, `AnthropicMessages.protocol`, `Gemini.protocol`, `BedrockConverse.protocol`.
|
||||
- **`Endpoint`** (`src/route/endpoint.ts`) — URL construction. The host, path, and route query live on the endpoint. `Endpoint.path("/chat/completions", { baseURL })` is the common case; pass a function for paths that embed the model id or a body field (e.g. `Endpoint.path(({ body }) => `/model/${body.modelId}/converse-stream`)`).
|
||||
- **`Auth`** (`src/route/auth.ts`) — per-request transport authentication. Provider facades configure credentials onto the route before model selection, usually via `Auth.bearer(apiKey)` or `Auth.header(name, apiKey)`. Routes that need per-request signing (Bedrock SigV4, future Vertex IAM, Azure AAD) implement `Auth` as a function that signs the body and merges signed headers into the result.
|
||||
- **`Framing`** (`src/route/framing.ts`) — bytes → frames. SSE (`Framing.sse`) is shared; Bedrock keeps its AWS event-stream framing as a typed `Framing<object>` value alongside its protocol.
|
||||
|
||||
Compose them via `Route.make(...)`:
|
||||
|
||||
```ts
|
||||
export const route = Route.make({
|
||||
id: "openai-chat",
|
||||
provider: "openai",
|
||||
protocol: OpenAIChat.protocol,
|
||||
endpoint: Endpoint.path("/chat/completions", {
|
||||
baseURL: "https://api.openai.com/v1",
|
||||
}),
|
||||
auth: Auth.bearer(),
|
||||
framing: Framing.sse,
|
||||
})
|
||||
```
|
||||
|
||||
Route defaults are request-shaping defaults such as `headers`, `limits`, `generation`, `providerOptions`, and `http`. Endpoint host/query belongs on the route endpoint. Selected `LanguageModel` values carry only model id, provider id, and the configured route value. Model capability/catalog metadata lives outside this package; protocol support is enforced by request lowering and typed `AIError`s.
|
||||
|
||||
The four-axis decomposition is the reason DeepSeek, TogetherAI, Cerebras, Baseten, Fireworks, and DeepInfra all reuse `OpenAIChat.protocol` verbatim — each provider deployment is a 5-15 line `Route.make(...)` call instead of a 300-400 line route clone. Bug fixes in one protocol propagate to every consumer of that protocol in a single commit.
|
||||
|
||||
When a provider supports multiple physical transports, selection remains execution policy below its semantic route. `OpenResponsesChannel.transport(...)` owns the provider-neutral Responses WebSocket concept: it prepares one final request, executes HTTP by default, strips WebSocket-disallowed fields, and passes a generic channel exchange to a per-call `WebSocketChannelExecutor` when supplied. Provider-specific Responses routes opt in with handshake and connection-age policy. `Route.streamPrepared` owns decoding and acknowledges channel completion only after successful full consumption.
|
||||
|
||||
### URL Construction
|
||||
|
||||
`Endpoint` owns `{ baseURL, path, query }`. Each protocol route includes a canonical endpoint when the provider has one (e.g. `https://api.openai.com/v1`); provider helpers override endpoint fields by configuring the route before selecting a model. Routes that have no canonical URL (OpenAI-compatible Chat, GitHub Copilot) require configuration before execution.
|
||||
|
||||
For providers where the URL is derived from typed inputs (Azure resource name, Bedrock region), the provider helper configures the route endpoint before calling `.model(...)`. Use `AtLeastOne<T>` from `route/auth-options.ts` for inputs that accept either of two derivation paths (Azure: `resourceName` or `baseURL`).
|
||||
|
||||
### Provider Facades
|
||||
|
||||
Provider-facing APIs are configured facades over route values. Endpoint/auth/resource/API-version setup happens before model selection, and model selectors accept only a model or deployment id:
|
||||
|
||||
```ts
|
||||
const openai = OpenAI.configure({ apiKey, baseURL })
|
||||
const model = openai.responses("gpt-4o-mini")
|
||||
|
||||
const azure = Azure.configure({ resourceName, apiKey, apiVersion: "v1" })
|
||||
const deployment = azure.responses("my-deployment")
|
||||
|
||||
const gateway = CloudflareAIGateway.configure({ accountId, gatewayId, gatewayApiKey, apiKey })
|
||||
const proxied = gateway.model("openai/gpt-4o-mini")
|
||||
```
|
||||
|
||||
Keep provider facades small and explicit:
|
||||
|
||||
- Use branded `ProviderID.make(...)` and `ModelID.make(...)` where ids are constructed directly.
|
||||
- Use `model` for the default API path and named methods for provider-native alternatives such as OpenAI `responses` and `chat`.
|
||||
- Put provider-specific setup on `.configure(...)`; do not add `model(id, overrides)` as a duplicate construction path.
|
||||
- Export lower-level `routes` arrays separately only when advanced internal wiring needs them.
|
||||
- Prefer `apiKey` as provider-specific sugar and `auth` as the explicit override; keep them mutually exclusive in provider option types with `ProviderAuthOption`.
|
||||
- Resolve `apiKey` → `Auth` with `AuthOptions.bearer(options, "<PROVIDER>_API_KEY")` (it honors an explicit `auth` override and falls back to `Auth.config(envVar)` so missing keys surface a typed `Authentication` error rather than a runtime crash).
|
||||
- Use separate top-level facades for products with different required setup, such as `CloudflareAIGateway` and `CloudflareWorkersAI`.
|
||||
|
||||
`Provider.make(...)` remains available for simple static provider definitions, but new built-in providers should prefer plain configured facades unless a helper removes real duplication without adding runtime behavior.
|
||||
|
||||
### Provider Package Entrypoints
|
||||
|
||||
Catalog-selected native providers use package-like export paths from `@opencode-ai/ai`. They are internal entrypoints in one npm package, not separately published provider packages. Every entrypoint implements `ProviderPackage.Definition` and exposes `model(modelID, settings)`, where settings are serializable provider configuration plus common `headers`, `body`, and `limits` overlays.
|
||||
|
||||
```ts
|
||||
import { model } from "@opencode-ai/ai/providers/openai/responses"
|
||||
|
||||
const selected = model("gpt-5", {
|
||||
apiKey,
|
||||
})
|
||||
```
|
||||
|
||||
Keep semantic APIs as separate entrypoints, such as OpenAI `chat` and `responses`. Transport is execution policy: OpenAI Responses uses HTTP by default and may receive a per-call WebSocket channel executor through `StreamOptions` without changing model or route identity.
|
||||
|
||||
Do not expose `Route` in provider package settings. Route composition stays an implementation detail behind `model(...)`.
|
||||
|
||||
### Folder layout
|
||||
|
||||
```
|
||||
packages/ai/src/
|
||||
schema/ canonical Schema model, split by concern
|
||||
ids.ts branded IDs, literal types, ProviderMetadata
|
||||
options.ts Generation/Provider/Http options, Limits, LanguageModel, cache policy
|
||||
messages.ts content parts, Message, ToolDefinition, LLMRequest
|
||||
events.ts Usage, individual events, LLMEvent, LLMResponse
|
||||
errors.ts error reasons, AIError, ToolFailure
|
||||
index.ts barrel
|
||||
llm.ts request constructors and convenience helpers
|
||||
route/
|
||||
index.ts @opencode-ai/ai/route advanced barrel
|
||||
client.ts Route.make + LLMClient.stream/generate
|
||||
executor.ts RequestExecutor service + transport error mapping
|
||||
protocol.ts Protocol type + Protocol.make
|
||||
endpoint.ts Endpoint type + Endpoint.path
|
||||
auth.ts Auth type + Auth.bearer / Auth.apiKeyHeader / Auth.passthrough
|
||||
auth-options.ts ProviderAuthOption shape, AuthOptions.bearer, AtLeastOne helper
|
||||
framing.ts Framing type + Framing.sse
|
||||
transport/ transport implementations
|
||||
index.ts Transport execution types + HttpTransport / WebSocketTransport namespaces
|
||||
websocket-channel.ts generic sequential channel executor/driver contract
|
||||
http.ts HttpTransport.httpJson — POST + framing
|
||||
websocket.ts direct one-request channel executor + raw socket adapter
|
||||
protocols/
|
||||
shared.ts ProviderShared toolkit used inside protocol impls
|
||||
openai-chat.ts protocol + route (compose OpenAIChat.protocol)
|
||||
open-responses.ts provider-neutral Responses protocol baseline
|
||||
open-responses-channel.ts provider-neutral Responses WebSocket transport factory
|
||||
openai-responses.ts OpenAI tools/events and channel policy composed over OpenResponses
|
||||
anthropic-messages.ts
|
||||
gemini.ts
|
||||
bedrock-converse.ts
|
||||
bedrock-event-stream.ts framing for AWS event-stream binary frames
|
||||
openai-compatible-chat.ts route that reuses OpenAIChat.protocol, no canonical URL
|
||||
openai-compatible-responses.ts deployment adapter that reuses OpenResponses.protocol, no canonical URL
|
||||
utils/ per-protocol helpers (auth, cache, media, tool-stream, ...)
|
||||
providers/
|
||||
openai-compatible.ts generic Chat helper + family model helpers
|
||||
openai-compatible-responses.ts generic Responses helper
|
||||
openai-compatible-profile.ts family defaults (deepseek, togetherai, ...)
|
||||
azure.ts / amazon-bedrock.ts / cloudflare.ts / github-copilot.ts / google.ts / xai.ts / openai.ts / anthropic.ts / openrouter.ts
|
||||
tool.ts typed tool() helper
|
||||
tool-runtime.ts narrow one-call typed tool dispatcher
|
||||
```
|
||||
|
||||
The dependency arrow points down: `providers/*.ts` files import protocol routes and auth-option utilities; protocol modules import `endpoint`, `auth`, `framing`, and transport pieces. Protocols do not import provider facades. Lower-level modules know nothing about provider catalog metadata. `OpenAIResponses` composes the provider-neutral `OpenResponses` protocol; the baseline never imports the OpenAI extension.
|
||||
|
||||
### Shared protocol helpers
|
||||
|
||||
`ProviderShared` exports a small toolkit used inside protocol implementations to keep them focused on provider-native shapes:
|
||||
|
||||
- `joinText(parts)` — joins an array of `TextPart` (or anything with a `.text`) with newlines. Use this anywhere a protocol flattens text content into a single string for a provider field.
|
||||
- `parseToolInput(route, name, raw)` — Schema-decodes a tool-call argument string with the canonical "Invalid JSON input for `<route>` tool call `<name>`" error message. Treats empty input as `{}`.
|
||||
- `parseJson(route, raw, message)` — generic JSON-via-Schema decode for non-tool bodies.
|
||||
- `eventError(route, message, ...)` — typed `InvalidProviderOutput` constructor for stream-time decode failures.
|
||||
- `validateWith(decoder)` — maps Schema decode errors to `InvalidRequest`. `Route.make(...)` uses this for body validation; lower-level routes can reuse it.
|
||||
- `matchToolChoice(provider, choice, branches)` — branches over `LLMRequest["toolChoice"]` for provider-specific lowering.
|
||||
|
||||
If you find yourself copying a 3-to-5-line snippet between two protocols, lift it into `ProviderShared` next to these helpers rather than duplicating.
|
||||
|
||||
### Chronological System Updates
|
||||
|
||||
`LLMRequest.system` is the initial privileged prompt that applies ahead of the conversation. `Message.system(...)` is a separate, provider-neutral chronological operator update inside `LLMRequest.messages`; it applies only from its position in history onward and accepts text content only.
|
||||
|
||||
Native chronological system messages are route/model-specific. Anthropic Messages lowers them natively for Claude Opus 4.8 (`claude-opus-4-8`). Other routes and models intentionally lower the update in place into ordinary user-compatible text using this stable escaped representation:
|
||||
|
||||
```text
|
||||
<system-update>
|
||||
...
|
||||
</system-update>
|
||||
```
|
||||
|
||||
The wrapped-user fallback preserves ordering while visibly lowering authority. Never silently pass a raw chronological `role: "system"` through a route that might reject it. Do not insert raw retrieved documents, tool output, or web content into privileged chronological system updates; keep untrusted content in ordinary user/tool channels.
|
||||
|
||||
### Tools
|
||||
|
||||
Tool loops are represented in common messages and events:
|
||||
|
||||
```ts
|
||||
const call = ToolCallPart.make({ id: "call_1", name: "lookup", input: { query: "weather" } })
|
||||
const result = Message.tool({ id: "call_1", name: "lookup", result: { forecast: "sunny" } })
|
||||
|
||||
const followUp = LLM.request({
|
||||
model,
|
||||
messages: [Message.user("Weather?"), Message.assistant([call]), result],
|
||||
})
|
||||
```
|
||||
|
||||
Routes lower these into provider-native assistant tool-call messages and tool-result messages. Streaming providers should emit `tool-input-delta` events while arguments arrive, then a final `tool-call` event with parsed input.
|
||||
|
||||
### Tool dispatch
|
||||
|
||||
`LLM.stream(request)` and `LLM.generate(request)` each run exactly one provider turn. Add tool schemas to `request.tools` with `Tool.toDefinitions(tools)`. When a caller wants the package's typed one-call execution behavior, pass each canonical local `tool-call` event to `ToolRuntime.dispatch(tools, call)`.
|
||||
|
||||
```ts
|
||||
const get_weather = tool({
|
||||
description: "Get current weather for a city",
|
||||
parameters: Schema.Struct({ city: Schema.String }),
|
||||
success: Schema.Struct({ temperature: Schema.Number, condition: Schema.String }),
|
||||
execute: ({ city }) =>
|
||||
Effect.gen(function* () {
|
||||
// city: string — typed from parameters Schema
|
||||
const data = yield* WeatherApi.fetch(city)
|
||||
return { temperature: data.temp, condition: data.cond }
|
||||
// return type checked against success Schema
|
||||
}),
|
||||
})
|
||||
|
||||
const tools = { get_weather, get_time, ... }
|
||||
const events = yield* LLM.stream(
|
||||
LLMRequest.update(request, { tools: Tool.toDefinitions(tools) }),
|
||||
).pipe(Stream.runCollect)
|
||||
|
||||
const call = Array.from(events).find(LLMEvent.is.toolCall)
|
||||
if (call && !call.providerExecuted) {
|
||||
const dispatched = yield* ToolRuntime.dispatch(tools, call)
|
||||
// Persist call + dispatched.result, then construct the next request explicitly.
|
||||
}
|
||||
```
|
||||
|
||||
The dispatcher:
|
||||
|
||||
- On `tool-call`: looks up the named tool, decodes input against `parameters` Schema, dispatches to the typed `execute`, encodes the result against `success` Schema, and returns canonical `tool-result` events.
|
||||
- Does not stream providers, construct Session events, schedule fibers, append history, count steps, or continue model rounds.
|
||||
- Leaves persistence and continuation to the enclosing product flow.
|
||||
|
||||
Handler dependencies (services, permissions, plugin hooks, abort handling) are closed over by the consumer at tool-construction time. Build the tools record inside an `Effect.gen` once and reuse it across many dispatches.
|
||||
|
||||
Errors must be expressed as `ToolFailure`. The runtime catches it and emits a `tool-error` event, then a `tool-result` of `type: "error"`, so the model can self-correct on the next step. Anything that is not a `ToolFailure` is treated as a defect and fails the stream. Three recoverable error paths produce `tool-error` events:
|
||||
|
||||
- The model called an unknown tool name.
|
||||
- Input failed the `parameters` Schema.
|
||||
- The handler returned a `ToolFailure`.
|
||||
|
||||
Provider-defined / hosted tools (Anthropic `web_search` / `code_execution` / `web_fetch`, OpenAI Responses `web_search_call` / `file_search_call` / `code_interpreter_call` / `mcp_call` / `local_shell_call` / `image_generation_call` / `computer_use_call`) pass through the runtime untouched:
|
||||
|
||||
- Routes surface the model's call as a `tool-call` event with `providerExecuted: true`, and the provider's result as a matching `tool-result` event with `providerExecuted: true`.
|
||||
- Callers detect `providerExecuted` on `tool-call` and **skip local dispatch** — no handler is invoked and no `tool-error` is raised for "unknown tool". The provider already executed it.
|
||||
- Callers that continue should retain both events in explicit history when the protocol requires it. Anthropic encodes them back as `server_tool_use` + `web_search_tool_result` (or `code_execution_tool_result` / `web_fetch_tool_result`) blocks; OpenAI Responses callers typically use `previous_response_id` instead of resending hosted-tool items.
|
||||
|
||||
Add provider-defined tools to `request.tools` (no runtime entry needed). The matching route must know how to lower the tool definition into the provider-native shape; right now Anthropic accepts `web_search` / `code_execution` / `web_fetch` and OpenAI Responses accepts the hosted tool names listed above.
|
||||
|
||||
## Protocol File Style
|
||||
|
||||
Protocol files should look self-similar. Provider quirks belong behind named helpers so a new route can be reviewed by comparing the same sections across files.
|
||||
|
||||
### Section order
|
||||
|
||||
Use this order for every protocol module:
|
||||
|
||||
1. Public model input
|
||||
2. Request body schema
|
||||
3. Streaming event schema
|
||||
4. Parser state
|
||||
5. Request body construction (`fromRequest`)
|
||||
6. Stream parsing (`step` and per-event handlers)
|
||||
7. Protocol and route
|
||||
8. Protocol route export
|
||||
|
||||
### Rules
|
||||
|
||||
- Keep protocol files focused on the protocol. Move provider-specific projection, signing, media normalization, or other bulky transformations into `src/protocols/utils/*`.
|
||||
- Use `Effect.fn("Provider.fromRequest")` for request body construction entrypoints. Use `Effect.fn(...)` for event handlers that yield effects; keep purely synchronous handlers as plain functions returning a `StepResult` that the dispatcher lifts via `Effect.succeed(...)`.
|
||||
- Parser state owns terminal information. The state machine records finish reason, usage, and pending tool calls; emit one terminal `finish` event (or `provider-error`) for each completed response. If a provider splits reason and usage across events, merge them in parser state before flushing.
|
||||
- Emit exactly one terminal `finish` event for a completed response, normally after a matching `step-finish`. Use `stream.terminal` to stop reading when the provider has a completion sentinel; use `stream.onHalt` when the final event must be flushed after the framed stream ends.
|
||||
- Use shared helpers for repeated protocol policy such as text joining, usage totals, JSON parsing, and tool-call accumulation. `ToolStream` (`protocols/utils/tool-stream.ts`) accumulates streamed tool-call arguments uniformly.
|
||||
- Make intentional provider differences explicit in helper names or comments. If two protocol files differ visually, the reason should be obvious from the names.
|
||||
- Prefer dispatched per-event handlers (`onMessageStart`, `onContentBlockDelta`, ...) called from a small top-level `step` switch over a long if-chain. The dispatcher keeps the event surface visible at a glance.
|
||||
- Keep tests in the same conceptual order as the protocol: basic prepare, tools prepare, unsupported lowering, text/usage parsing, tool streaming, finish reasons, provider errors.
|
||||
|
||||
### Review checklist
|
||||
|
||||
- Can the file be skimmed side-by-side with `openai-chat.ts` without hunting for equivalent sections?
|
||||
- Are provider quirks named, isolated, and covered by focused tests?
|
||||
- Does request body construction validate unsupported common content at the protocol boundary?
|
||||
- Does stream parsing emit stable common events without leaking provider event order to callers?
|
||||
- Does `toolChoice: "none"` behavior read as intentional?
|
||||
|
||||
## Recording Tests
|
||||
|
||||
Recorded tests use one cassette file per scenario. A cassette holds an ordered array of `{ request, response }` interactions, so multi-step flows (tool loops, retries, polling) record into a single file. Use `recordedTests({ prefix, requires })` and let the helper derive cassette names from test names:
|
||||
|
||||
```ts
|
||||
const recorded = recordedTests({ prefix: "openai-chat", requires: ["OPENAI_API_KEY"] })
|
||||
|
||||
recorded.effect("streams text", () =>
|
||||
Effect.gen(function* () {
|
||||
// test body
|
||||
}),
|
||||
)
|
||||
```
|
||||
|
||||
Replay is the default. `RECORD=true` records fresh cassettes locally and requires the listed env vars; unset `CI` before recording because CI always forces replay. Cassettes are written as pretty-printed JSON so multi-interaction diffs stay reviewable.
|
||||
|
||||
Pass `provider`, `protocol`, and optional `tags` to `recordedTests(...)` / `recorded.effect.with(...)` so cassettes carry searchable metadata. Use recorded-test filters to replay or record a narrow subset without rewriting a whole file:
|
||||
|
||||
- `RECORDED_PROVIDER=openai` matches tests tagged with `provider:openai`; comma-separated values are allowed.
|
||||
- `RECORDED_PREFIX=openai-chat` matches cassette groups by `recordedTests({ prefix })`; comma-separated values are allowed.
|
||||
- `RECORDED_TAGS=tool` requires all listed tags to be present, e.g. `RECORDED_TAGS=provider:togetherai,tool`.
|
||||
- `RECORDED_TEST="streams text"` matches by test name, kebab-case test id, or cassette path.
|
||||
|
||||
Filters apply in replay and record mode. Combine them with `RECORD=true` when refreshing only one provider or scenario.
|
||||
|
||||
**Binary response bodies.** Most providers stream text (SSE, JSON). The recorder treats known textual media types (`text/*`, JSON/XML structured types, JavaScript, forms, YAML, and SVG) as text and stores every other response as base64 with `bodyEncoding: "base64"`. This preserves binary formats such as AWS event-stream frames without a lossy UTF-8 round trip.
|
||||
|
||||
**Matching strategy.** A runtime request atomically claims the first unused recorded interaction that matches its method, URL, allow-listed headers, and canonical JSON body. Distinct requests may replay in any order or concurrently. Repeated identical requests consume their matching responses in cassette order, preserving deterministic retry and polling behavior. `scriptedResponses` (in `test/lib/http.ts`) is the deterministic counterpart for tests that don't need a live provider; it scripts response bodies in order without reading from disk.
|
||||
|
||||
Do not blanket re-record an entire test file when adding one cassette. `RECORD=true` rewrites every recorded case that runs, and provider streams contain volatile IDs, timestamps, fingerprints, and obfuscation fields. Prefer deleting the one cassette you intend to refresh, or run a focused test pattern that only registers the scenario you want to record. Keep stable existing cassettes unchanged unless their request shape or expected behavior changed.
|
||||
@@ -1,392 +0,0 @@
|
||||
# @opencode-ai/ai
|
||||
|
||||
Schema-first AI primitives for opencode. Provider quirks live in adapters, not in calling code.
|
||||
|
||||
```ts
|
||||
import { Effect, Layer } from "effect"
|
||||
import { LLM, LLMClient } from "@opencode-ai/ai"
|
||||
import { RequestExecutor } from "@opencode-ai/ai/route"
|
||||
import { OpenAI } from "@opencode-ai/ai/providers"
|
||||
|
||||
const model = OpenAI.configure({ apiKey: process.env.OPENAI_API_KEY }).responses("gpt-4o-mini")
|
||||
|
||||
const request = LLM.request({
|
||||
model,
|
||||
system: "You are concise.",
|
||||
prompt: "Say hello in one short sentence.",
|
||||
generation: { maxTokens: 40 },
|
||||
})
|
||||
|
||||
const program = Effect.gen(function* () {
|
||||
const response = yield* LLMClient.generate(request)
|
||||
console.log(response.text)
|
||||
})
|
||||
|
||||
const llmLayer = LLMClient.layer.pipe(Layer.provide(RequestExecutor.fetchLayer))
|
||||
|
||||
await Effect.runPromise(program.pipe(Effect.provide(llmLayer)))
|
||||
```
|
||||
|
||||
Run `LLMClient.stream(request)` instead of `generate` when you want incremental `LLMEvent`s. The event stream is provider-neutral — same shape across OpenAI Chat, OpenAI Responses, Anthropic Messages, Gemini, Bedrock Converse, and any OpenAI-compatible deployment.
|
||||
|
||||
## Image generation
|
||||
|
||||
Use `Image.generate` with an image model for direct asset generation:
|
||||
|
||||
```ts
|
||||
import { Image, ImageInput } from "@opencode-ai/ai"
|
||||
import { OpenAI } from "@opencode-ai/ai/providers"
|
||||
|
||||
const program = Effect.gen(function* () {
|
||||
const response = yield* Image.generate({
|
||||
model: OpenAI.configure({ apiKey: process.env.OPENAI_API_KEY }).image("gpt-image-2"),
|
||||
prompt: "A robot tending a rooftop garden",
|
||||
options: {
|
||||
n: 2,
|
||||
size: "1024x1024",
|
||||
quality: "high", // inferred from the OpenAI image model
|
||||
outputFormat: "webp",
|
||||
future_option: true, // unknown native options pass through unchanged
|
||||
},
|
||||
})
|
||||
|
||||
return response.images // GeneratedImage[] with owned bytes or a provider URL
|
||||
})
|
||||
```
|
||||
|
||||
Pass ordered image inputs to the same method for editing, composition, or image-conditioned generation:
|
||||
|
||||
```ts
|
||||
const response =
|
||||
yield *
|
||||
Image.generate({
|
||||
model,
|
||||
prompt: "Combine these product photos into one studio scene",
|
||||
images: [
|
||||
ImageInput.bytes(firstBytes, "image/png"),
|
||||
ImageInput.url("https://example.com/second.webp"),
|
||||
ImageInput.file("file_123"),
|
||||
],
|
||||
options,
|
||||
http,
|
||||
})
|
||||
```
|
||||
|
||||
`ImageInput.fileUri(uri, mediaType)` represents provider file URIs such as Gemini Files. Raw strings are not
|
||||
accepted as image inputs, avoiding ambiguity between base64, URLs, and provider IDs. Empty or omitted `images`
|
||||
uses text-to-image generation; a non-empty array selects the provider's edit behavior without enforcing provider
|
||||
image-count limits locally. `images` is the only common image-editing field. OpenAI uses multipart for byte/data-URL
|
||||
edits and its JSON reference body for URL or file-ID edits. Its provider-specific `options.mask` accepts an
|
||||
`ImageInput` for inpainting:
|
||||
|
||||
```ts
|
||||
yield *
|
||||
Image.generate({
|
||||
model: OpenAI.configure({ apiKey }).image("gpt-image-2"),
|
||||
prompt,
|
||||
images: [ImageInput.bytes(sourceBytes, "image/png")],
|
||||
options: { mask: ImageInput.bytes(maskBytes, "image/png") },
|
||||
})
|
||||
```
|
||||
|
||||
The OpenAI adapter extracts this helper value into the edit request's native `mask` field rather than passing the
|
||||
tagged `ImageInput` object through as an ordinary option. On multipart requests, `http.body` can override option
|
||||
fields but not structural `model`, `prompt`, `image[]`, or `mask` fields, and the transport owns the multipart
|
||||
`Content-Type` boundary. For JSON requests, `http.body` remains the final raw-native overlay. Gemini does not fetch
|
||||
public HTTP URLs, and hosted Z.ai image generation does not accept image inputs. These cases fail with
|
||||
`InvalidRequest` before network I/O.
|
||||
|
||||
Provider-native image options belong to each request. Raw `http.body` fields have final precedence over them:
|
||||
|
||||
```ts
|
||||
const model = OpenAI.configure({ apiKey }).image("gpt-image-2")
|
||||
|
||||
yield *
|
||||
Image.generate({
|
||||
model,
|
||||
prompt,
|
||||
options: { quality: "medium" },
|
||||
http,
|
||||
})
|
||||
```
|
||||
|
||||
xAI image models use the same request API with xAI-native controls:
|
||||
|
||||
```ts
|
||||
yield *
|
||||
Image.generate({
|
||||
model: XAI.configure({ apiKey }).image("any-model-id"),
|
||||
prompt,
|
||||
options: {
|
||||
n: 2,
|
||||
aspectRatio: "16:9",
|
||||
resolution: "1k",
|
||||
responseFormat: "b64_json",
|
||||
future_option: true,
|
||||
},
|
||||
http,
|
||||
})
|
||||
```
|
||||
|
||||
Google's current Gemini image models use the same direct API:
|
||||
|
||||
```ts
|
||||
import { Google } from "@opencode-ai/ai/providers"
|
||||
|
||||
const googleProgram = Effect.gen(function* () {
|
||||
const response = yield* Image.generate({
|
||||
model: Google.configure({ apiKey }).image("any-model-id"),
|
||||
prompt: "A robot tending a rooftop garden",
|
||||
options: {
|
||||
aspectRatio: "16:9",
|
||||
imageSize: "2K",
|
||||
seed: 42,
|
||||
thinkingLevel: "HIGH",
|
||||
includeThoughts: true,
|
||||
futureOption: true,
|
||||
},
|
||||
http,
|
||||
})
|
||||
|
||||
return response.images
|
||||
})
|
||||
```
|
||||
|
||||
Google image options are request-scoped and inferred from the selected model. Known fields autocomplete while
|
||||
future string values and arbitrary native Gemini `generationConfig` fields remain available. Native fields override
|
||||
their mapped aliases, and `http.body` is the final deep overlay. The selected model ID is sent to Gemini
|
||||
`generateContent` without a local allowlist.
|
||||
|
||||
Z.ai image models infer open Z.ai-native options from the selected model:
|
||||
|
||||
```ts
|
||||
yield *
|
||||
Image.generate({
|
||||
model: ZAI.configure({ apiKey }).image("any-model-id"),
|
||||
prompt,
|
||||
options: {
|
||||
quality: "hd",
|
||||
userID: "user-123",
|
||||
future_option: true,
|
||||
},
|
||||
http,
|
||||
})
|
||||
```
|
||||
|
||||
Z.ai does not include trustworthy MIME metadata for output URLs, so generated images use
|
||||
`application/octet-stream`. Output URLs expire after 30 days; download and persist them promptly if they must
|
||||
remain available.
|
||||
|
||||
Conversational image generation remains part of the LLM interaction. OpenAI Responses exposes it through its hosted image tool:
|
||||
|
||||
```ts
|
||||
const program = Effect.gen(function* () {
|
||||
const response = yield* LLM.generate(
|
||||
LLM.request({
|
||||
model: OpenAI.configure({ apiKey }).responses("gpt-5"),
|
||||
prompt: "Design a solarpunk rooftop garden, then show me.",
|
||||
tools: [OpenAI.imageGeneration({ quality: "high" })],
|
||||
}),
|
||||
)
|
||||
|
||||
return response.message
|
||||
})
|
||||
```
|
||||
|
||||
The hosted result is represented as a provider-executed tool call and tool result. Its image is a `file` content item with a data URI, so retaining `response.message` preserves the generated image for continuation.
|
||||
|
||||
## Public API
|
||||
|
||||
- **`LLM.request({...})`** — build a provider-neutral `LLMRequest`. Accepts ergonomic inputs (`system: string`, `prompt: string`) that normalize into the canonical Schema classes.
|
||||
- **`LLM.generate` / `LLM.stream`** — re-exported from `LLMClient` for one-import use.
|
||||
- **`Message.user(...)` / `Message.assistant(...)` / `Message.tool(...)`** — message constructors from the canonical schema model.
|
||||
- **`LanguageModel.make(...)` / `ToolCallPart.make(...)` / `ToolResultPart.make(...)` / `ToolDefinition.make(...)`** — model and tool-related constructors from the canonical schema model.
|
||||
- **`LLMEvent.is.*`** — typed guards (`is.textDelta`, `is.toolCall`, `is.finish`, …) for filtering streams.
|
||||
- **`Image.generate({...})`** — generate images through a provider-neutral image request and response model.
|
||||
- **`ImageClient`** — Effect service and layer for image execution, parallel to `LLMClient`.
|
||||
|
||||
## Testing
|
||||
|
||||
Use the deterministic test client from `@opencode-ai/ai/testing` to script provider-neutral responses and inspect
|
||||
the requests sent by code under test:
|
||||
|
||||
```ts
|
||||
import { Effect } from "effect"
|
||||
import { TestLLM } from "@opencode-ai/ai/testing"
|
||||
|
||||
const testLLM = TestLLM.layer({
|
||||
fallback: TestLLM.text("Hello from the test model", "text-1"),
|
||||
})
|
||||
|
||||
// TestLLM.clientLayer provides LLMClient.Service and consumes TestLLM.Service.
|
||||
const programWithTestClient = Effect.gen(function* () {
|
||||
const result = yield* program
|
||||
const test = yield* TestLLM.Service
|
||||
console.log(test.requests)
|
||||
return result
|
||||
}).pipe(Effect.provide(TestLLM.clientLayer), Effect.provide(testLLM))
|
||||
```
|
||||
|
||||
`TestLLM.push(...)` scripts one-shot responses, `TestLLM.always(...)` changes the fallback, and
|
||||
`TestLLM.wait(...)` lets concurrent tests wait until a request has arrived. Every received canonical request is
|
||||
available on the yielded `TestLLM.Service`.
|
||||
|
||||
## Caching
|
||||
|
||||
Prompt caching is **on by default**. Every `LLMRequest` resolves to `cache: "auto"` unless the caller opts out with `cache: "none"`. Each protocol translates `CacheHint`s to its wire format (`cache_control` on Anthropic, `cachePoint` on Bedrock; OpenAI and Gemini do implicit caching server-side and don't need inline markers — auto is a no-op there).
|
||||
|
||||
### Auto placement
|
||||
|
||||
`"auto"` places up to four breakpoints — the last tool definition, the first system part, the last system part when distinct, and the final message boundary. These expose successively larger reusable prefixes for tools, the base agent, project instructions, and the active conversation. The rolling final-message boundary is the load-bearing detail in tool loops: it advances on every request so the previous cache entry stays within Anthropic's 20-block lookback.
|
||||
|
||||
Tools precede every system and conversation block in the provider prefix, so tool definitions must remain byte-stable and deterministically ordered for downstream breakpoints to remain reusable.
|
||||
|
||||
The math justifies the default: Anthropic's 5-minute cache write is 1.25× base, read is 0.1×, so a single reuse within 5 minutes already wins. One-shot completions below the per-model minimum-cacheable-token threshold silently no-op on the wire, so the worst case is harmless.
|
||||
|
||||
### Opting out
|
||||
|
||||
```ts
|
||||
LLM.request({
|
||||
model,
|
||||
system,
|
||||
prompt: "one-off question",
|
||||
cache: "none",
|
||||
})
|
||||
```
|
||||
|
||||
### Granular policy
|
||||
|
||||
```ts
|
||||
cache: {
|
||||
tools?: boolean,
|
||||
system?: boolean,
|
||||
messages?: "latest-user-message" | "latest-assistant" | { tail: number },
|
||||
ttlSeconds?: number, // ≥ 3600 → 1h on Anthropic/Bedrock; else 5m
|
||||
}
|
||||
```
|
||||
|
||||
### Manual hints
|
||||
|
||||
Inline `CacheHint` on any text / system / tool / tool-result part overrides automatic placement. The auto policy preserves manual hints, counts them against Anthropic and Bedrock's four-breakpoint limit, and only fills the remaining slots.
|
||||
|
||||
```ts
|
||||
LLM.request({
|
||||
model,
|
||||
system: [
|
||||
{ type: "text", text: "stable system prompt", cache: { type: "ephemeral" } },
|
||||
],
|
||||
...
|
||||
})
|
||||
```
|
||||
|
||||
### Provider behavior table
|
||||
|
||||
| Protocol | `cache: "auto"` |
|
||||
| ----------------------- | ------------------------------------------------------------------------- |
|
||||
| Anthropic Messages | emits up to 4 `cache_control` markers (4-breakpoint cap enforced) |
|
||||
| Bedrock Converse | emits up to 4 `cachePoint` blocks (4-breakpoint cap enforced) |
|
||||
| OpenAI Chat / Responses | no-op (implicit caching above 1024 tokens) |
|
||||
| Gemini | no-op (implicit caching on 2.5+; explicit `CachedContent` is out-of-band) |
|
||||
|
||||
Normalized cache usage is read back into `response.usage.cacheReadInputTokens` and `cacheWriteInputTokens` across every provider.
|
||||
|
||||
## Providers
|
||||
|
||||
Provider facades configure endpoint/auth/deployment details first, then expose model selectors that take only a model or deployment id. The selected model carries the executable route value used at runtime.
|
||||
|
||||
```ts
|
||||
import { OpenAI, CloudflareAIGateway } from "@opencode-ai/ai/providers"
|
||||
|
||||
const openai = OpenAI.configure({ apiKey: process.env.OPENAI_API_KEY }).responses("gpt-4o-mini")
|
||||
const gateway = CloudflareAIGateway.configure({
|
||||
accountId: process.env.CLOUDFLARE_ACCOUNT_ID,
|
||||
gatewayApiKey: process.env.CLOUDFLARE_API_TOKEN,
|
||||
}).model("workers-ai/@cf/meta/llama-3.1-8b-instruct")
|
||||
```
|
||||
|
||||
Included providers: OpenAI, Anthropic, Google (Gemini), Google Vertex Gemini and Anthropic, Amazon Bedrock, Azure OpenAI, Cloudflare AI Gateway, Cloudflare Workers AI, GitHub Copilot, OpenRouter, xAI, Z.ai, plus generic OpenAI-compatible Chat and Responses entrypoints and an Anthropic Messages-compatible entrypoint.
|
||||
|
||||
### Package-like entrypoints
|
||||
|
||||
Native catalog integrations load provider behavior through package-like entrypoints. These are export paths from the same `@opencode-ai/ai` npm package, not independently published packages. Each entrypoint exports the same `model(modelID, settings)` contract, and `settings` contains serializable provider configuration plus common `headers`, `body`, and `limits` overlays.
|
||||
|
||||
```ts
|
||||
import { model } from "@opencode-ai/ai/providers/openai/responses"
|
||||
|
||||
const selected = model("gpt-5", {
|
||||
apiKey: process.env.OPENAI_API_KEY,
|
||||
headers: { "x-application": "opencode" },
|
||||
limits: { context: 200_000, output: 64_000 },
|
||||
})
|
||||
```
|
||||
|
||||
OpenAI Chat and OpenAI Responses are separate semantic entrypoints:
|
||||
|
||||
- `@opencode-ai/ai/providers/openai/chat`
|
||||
- `@opencode-ai/ai/providers/openai/responses`
|
||||
- `@opencode-ai/ai/providers/openai-compatible/responses`
|
||||
- `@opencode-ai/ai/providers/anthropic-compatible`
|
||||
- `@opencode-ai/ai/providers/google-vertex/gemini`
|
||||
- `@opencode-ai/ai/providers/google-vertex/chat`
|
||||
- `@opencode-ai/ai/providers/google-vertex/responses`
|
||||
- `@opencode-ai/ai/providers/google-vertex/messages`
|
||||
|
||||
OpenAI Responses has one semantic route and uses HTTP by default. Advanced callers may supply a per-call WebSocket channel executor through `StreamOptions`; transport policy does not change provider settings, model identity, or route identity. The provider-neutral Open Responses implementation owns the reusable WebSocket request and event contract, while each provider opts in with its own handshake and connection policy. Azure follows the same Chat/Responses split at `providers/azure/chat` and `providers/azure/responses`. Generic OpenAI-compatible Chat remains at `providers/openai-compatible`; the Responses adapter at `providers/openai-compatible/responses` uses the provider-neutral Open Responses protocol. OpenAI Responses extends that baseline with OpenAI tools, event variants, metadata, and defaults. Generic Anthropic Messages-compatible providers use `providers/anthropic-compatible`, which the named Anthropic provider composes. Google Gemini and Amazon Bedrock expose their single native API through their existing provider paths.
|
||||
|
||||
Vertex Gemini, Vertex Chat, Vertex Responses, and Vertex Messages are separate API entrypoints. All accept `project`, `location`, and an optional `accessToken`; when no explicit token or auth override is supplied they lazily use Google Application Default Credentials. Vertex Gemini instead selects express mode when `apiKey` or `GOOGLE_VERTEX_API_KEY` is present. Vertex Chat targets MaaS models through the OpenAI-compatible Chat Completions endpoint, while Vertex Responses targets Grok models and defaults `store` to `false` as required by Vertex. `providers/google-vertex` remains the default alias for `providers/google-vertex/gemini`.
|
||||
|
||||
Tuned Vertex Gemini deployments use model ids shaped like `endpoints/1234567890` and require OAuth or ADC; Vertex express-mode API keys support publisher models only.
|
||||
|
||||
```ts
|
||||
import { model } from "@opencode-ai/ai/providers/google-vertex/gemini"
|
||||
|
||||
model("gemini-3.5-flash", { project: "my-project", location: "global" })
|
||||
```
|
||||
|
||||
```ts
|
||||
import { model } from "@opencode-ai/ai/providers/google-vertex/chat"
|
||||
|
||||
model("deepseek-ai/deepseek-v3.2-maas", { project: "my-project", location: "global" })
|
||||
```
|
||||
|
||||
```ts
|
||||
import { model } from "@opencode-ai/ai/providers/google-vertex/responses"
|
||||
|
||||
model("xai/grok-4.20-reasoning", { project: "my-project", location: "global" })
|
||||
```
|
||||
|
||||
```ts
|
||||
import { model } from "@opencode-ai/ai/providers/google-vertex/messages"
|
||||
|
||||
model("claude-sonnet-4-6", { project: "my-project", location: "global" })
|
||||
```
|
||||
|
||||
Provider facades such as `OpenAI.configure(...).responses(...)` remain the direct application API. Package-like entrypoints are the self-similar loading contract used when a catalog selects behavior by export path.
|
||||
|
||||
Other provider exports listed above remain direct facades until they explicitly implement the package-like contract. Exporting a provider facade does not implicitly make it a catalog-loadable provider package.
|
||||
|
||||
## Provider options & HTTP overlays
|
||||
|
||||
Request options in order of stability:
|
||||
|
||||
1. **`generation`** — portable knobs (`maxTokens`, `temperature`, `topP`, `topK`, penalties, seed, stop).
|
||||
2. **`promptCacheKey`** — stable cache affinity lowered by every protocol that supports it.
|
||||
3. **`providerOptions: { <provider>: {...} }`** — typed-at-the-facade provider-specific knobs (OpenAI `store`, Anthropic `thinking`, Gemini `thinkingConfig`, OpenRouter routing).
|
||||
4. **`http: { body, headers, query }`** — last-resort serializable overlays merged into the final HTTP request. Reach for this only when a stable typed path doesn't yet exist.
|
||||
|
||||
Route/provider defaults are overridden by request-level values for each axis.
|
||||
|
||||
## Routes
|
||||
|
||||
Adding a new model or deployment is usually 5-15 lines using `Route.make({ protocol, endpoint, auth, framing, ... })`. The route owns endpoint/auth/framing and the protocol owns body construction plus stream parsing. Transports are reusable IO templates that receive route endpoint/auth at compile time. Capability/catalog metadata lives outside this low-level package; unsupported request shapes fail during protocol lowering. See `AGENTS.md` for the architectural detail.
|
||||
|
||||
## Effect
|
||||
|
||||
This package is built on Effect. Public methods return `Effect` or `Stream`; provide `LLMClient.layer` for LLM dispatch and `ImageClient.layer` for image dispatch, then import the provider/protocol modules for the routes you use. The example at `example/tutorial.ts` is a runnable walkthrough.
|
||||
|
||||
## See also
|
||||
|
||||
- `AGENTS.md` — architecture, route construction, contributor guide
|
||||
- `STATUS.md` — native provider parity status and AI SDK migration gaps
|
||||
- `example/tutorial.ts` — runnable end-to-end walkthrough
|
||||
- `test/provider/*.test.ts` — fixture-first protocol tests; `*.recorded.test.ts` files cover live cassettes
|
||||
@@ -1,107 +0,0 @@
|
||||
# LLM Provider Parity Status
|
||||
|
||||
Last reviewed: 2026-08-07
|
||||
|
||||
This file tracks the gap between the native `@opencode-ai/ai` package and the AI SDK provider packages that opencode still depends on for many catalog/runtime paths.
|
||||
|
||||
## Existing Status Sources
|
||||
|
||||
| File | What it tracks | Limitation |
|
||||
| ----------------------------------- | ----------------------------------------------------------------------- | ------------------------------------------------------- |
|
||||
| `packages/ai/DESIGN.md` | Future clean-break API proposal for `@opencode-ai/ai`. | Not a provider parity tracker. |
|
||||
| `packages/ai/example/call-sites.md` | Route/value/provider-facade migration checklist and call-site sketches. | Architecture migration only; not AI SDK package parity. |
|
||||
|
||||
## Current Implementation Snapshot
|
||||
|
||||
| Native slice | Source | Current state | Main gaps |
|
||||
| ---------------------------------- | --------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||||
| OpenAI Chat | `src/protocols/openai-chat.ts`, `src/providers/openai.ts` | Usable. Streams text, reasoning deltas, tool calls, usage, images, and common generation controls. | No typed structured-output / `response_format` path. Limited typed OpenAI option surface compared with SDK escape hatches. |
|
||||
| OpenAI Responses | `src/protocols/open-responses.ts`, `src/protocols/openai-responses.ts`, `src/providers/openai.ts` | Usable over HTTP by default, with optional per-call WebSocket channel execution on the same model and route identity. | No incremental `previous_response_id` path or persistent Session channel manager yet. Typed options cover only a subset of Responses fields. Structured output is still mostly synthetic-tool based. |
|
||||
| OpenAI-compatible Chat | `src/protocols/openai-compatible-chat.ts`, `src/providers/openai-compatible.ts` | Usable for generic Chat and several profiles: Baseten, Cerebras, DeepInfra, DeepSeek, Fireworks, Groq, TogetherAI. | Family quirks are mostly endpoint defaults, not full typed behavior. |
|
||||
| Open Responses-compatible | `src/protocols/open-responses.ts`, `src/protocols/openai-compatible-responses.ts`, `src/providers/openai-compatible-responses.ts` | Usable for deployments that implement the provider-neutral Open Responses protocol. The deployment adapter does not inherit OpenAI tools, events, metadata, or defaults. | No named family profiles or recorded deployment coverage yet. |
|
||||
| Anthropic-compatible Messages | `src/protocols/anthropic-messages.ts`, `src/providers/anthropic-compatible.ts` | Usable for deployments that implement the Anthropic Messages wire protocol. Named Anthropic composes this base; MiniMax M3 has recorded text and tool-loop coverage. | No named compatible family profiles yet. |
|
||||
| Anthropic Messages | `src/protocols/anthropic-messages.ts`, `src/providers/anthropic.ts` | Usable. Supports tools, thinking, cache control, images, server-hosted tool events, and usage. | Provider option surface is small. Beta/header handling, metadata, and newer Messages fields need a typed parity pass. |
|
||||
| Gemini Developer API | `src/protocols/gemini.ts`, `src/providers/google.ts` | Usable for Google API key flow. Supports text, images, tools, thinking signatures, and cache usage. | This is not Vertex. Typed provider options are narrow; many Gemini request fields currently require raw `http.body` overlays. |
|
||||
| Vertex Gemini | `src/protocols/gemini.ts`, `src/providers/google-vertex.ts` | Usable through API-key express mode, explicit OAuth tokens, or ADC with project/location endpoint derivation, including tuned `endpoints/...` deployments. | Core runner/catalog mapping and recorded provider coverage are missing. |
|
||||
| Vertex Chat | `src/protocols/openai-chat.ts`, `src/providers/google-vertex-chat.ts` | Usable for MaaS models through OpenAI-compatible Chat Completions with explicit OAuth tokens or ADC and project/location endpoint derivation. | Core runner/catalog mapping and recorded provider coverage are missing; MaaS family-specific request parity needs review. |
|
||||
| Vertex Responses | `src/protocols/open-responses.ts`, `src/providers/google-vertex-responses.ts` | Usable for Grok models through Open Responses with explicit OAuth tokens or ADC, project/location endpoint derivation, and an explicit `store: false` Vertex default. | Core runner/catalog mapping and recorded provider coverage are missing; stateful continuation is not supported by Vertex. |
|
||||
| Vertex Messages | `src/protocols/anthropic-messages.ts`, `src/providers/google-vertex-messages.ts` | Usable through explicit OAuth tokens or ADC, including global, regional, and `eu`/`us` multi-region endpoints. | Core runner/catalog mapping and recorded provider coverage are missing; Vertex-specific hosted-tool parity needs review. |
|
||||
| Bedrock Converse | `src/protocols/bedrock-converse.ts`, `src/providers/amazon-bedrock.ts` | Partial but real. Supports AWS event-stream framing, SigV4 with supplied credentials, bearer auth, tools, reasoning signatures, media, cache points, and recorded tests. | Native facade does not mirror the AI SDK plugin's default AWS credential chain/profile behavior. Runner/catalog mapping is missing. Guardrails, inference profiles, region-specific model ID fixes, and model-specific request fields need a parity pass. |
|
||||
| Azure OpenAI | `src/providers/azure.ts` using OpenAI Chat/Responses protocols | Partial. Supports resource/base URL setup, API key auth, API version query, Chat, and Responses selectors. | Core runner does not map `@ai-sdk/azure` to this native facade. AAD/token auth and Azure-specific endpoint variants need review. |
|
||||
| Cloudflare AI Gateway / Workers AI | `src/providers/cloudflare.ts` | Present via OpenAI-compatible Chat routes. | Useful but not part of the critical AI SDK replacement set yet. Needs per-product recorded coverage before relying on it broadly. |
|
||||
| OpenRouter | `src/providers/openrouter.ts` | Present with OpenRouter-specific usage/reasoning/prompt-cache options over Chat. | Responses-style OpenRouter support is absent. |
|
||||
| xAI | `src/providers/xai.ts` | Present with Responses and Chat selectors. | Needs package-parity review against the AI SDK xAI provider. |
|
||||
| GitHub Copilot | `src/providers/github-copilot.ts` | Present as explicit-base-URL OpenAI Chat/Responses facade. | Runtime/catalog integration remains specialized and should stay separate from public OpenAI-compatible defaults. |
|
||||
|
||||
## V2 Runner Status
|
||||
|
||||
`packages/core/src/session/runner/model.ts` currently resolves only this native subset from catalog `aisdk` metadata:
|
||||
|
||||
| Catalog API | Native route used today |
|
||||
| --------------------------------------------------- | ---------------------------- |
|
||||
| `aisdk:@ai-sdk/openai` | `OpenAIResponses.route` |
|
||||
| `aisdk:@ai-sdk/anthropic` | `AnthropicMessages.route` |
|
||||
| `aisdk:@ai-sdk/openai-compatible` with explicit URL | `OpenAICompatibleChat.route` |
|
||||
|
||||
Other `aisdk:` packages, including Google Vertex, Azure, and Bedrock, currently fall back through the AI SDK loader in the production runner. The dependency-free resolver seam rejects them with `SessionRunnerModel.UnsupportedPackageError`; they are not native route mappings yet.
|
||||
|
||||
## AI SDK Package Parity Matrix
|
||||
|
||||
| AI SDK package | Intended native target | Status | Biggest gaps |
|
||||
| --------------------------------- | --------------------------------------------------------- | ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||||
| `@ai-sdk/openai` | `OpenAI.chat`, `OpenAI.responses` | Partial / usable | Add complete typed option coverage, structured output strategy, explicit Responses continuation support, and runner execution policy for optional WebSocket channels. |
|
||||
| `@ai-sdk/openai-compatible` | Generic OpenAI-compatible Chat and Responses | Partial / usable | Decide per-family namespace/profile behavior and runner API selection for providers that support Responses versus Chat only. |
|
||||
| `@ai-sdk/anthropic` | `AnthropicMessages` | Partial / usable | Finish Messages API parity for headers/betas/metadata/newer fields and document hosted-tool continuation expectations. |
|
||||
| `@ai-sdk/google` | Gemini Developer API | Partial / usable | Add typed options for safety, response schema/modalities, cached content, grounding/search/code execution, and non-text output modes where supported. |
|
||||
| `@ai-sdk/google-vertex` | Vertex Gemini namespace/facade | Partial / usable | Add runner/catalog mapping, recorded coverage, and broader provider-option parity. |
|
||||
| `@ai-sdk/google-vertex/anthropic` | Anthropic Messages over Vertex namespace/facade | Partial / usable | Add runner/catalog mapping, recorded coverage, and Vertex-specific hosted-tool parity. |
|
||||
| `@ai-sdk/google-vertex/maas` | Vertex Chat | Partial / usable | Add runner/catalog mapping, recorded coverage, and MaaS family-specific request parity. |
|
||||
| `@ai-sdk/google-vertex/xai` | Vertex Chat / Responses | Partial / usable | Decide Chat/Responses selection for catalog models, add runner mapping and recorded coverage, and review xAI-specific request options. |
|
||||
| `@ai-sdk/azure` | Azure OpenAI Chat/Responses facade | Partial | Map runner/catalog metadata to native Azure, handle resourceName/baseURL/apiVersion variants, add AAD/token auth story, and verify Chat vs Responses deployment selection. |
|
||||
| `@ai-sdk/amazon-bedrock` | Bedrock Converse | Partial | Add default AWS credential chain/profile support, region/inference-profile model ID handling, provider option parity via `additionalModelRequestFields`, guardrails/performance config, and runner/catalog mapping. |
|
||||
| `@ai-sdk/amazon-bedrock/mantle` | Bedrock Mantle OpenAI-compatible Chat/Responses namespace | Partial / usable | Add default AWS credential chain/profile support; native catalog mapping currently requires bearer auth or explicit static credentials. |
|
||||
|
||||
## Highest-Risk Gaps
|
||||
|
||||
1. Runner support is narrower than the LLM package. The package has native provider facades for Google, Azure, and Bedrock, but the V2 Session runner only maps OpenAI, Anthropic, and explicit OpenAI-compatible Chat from `aisdk` catalog metadata.
|
||||
2. The Open Responses adapter is available through a separate package entrypoint, but the V2 runner still maps `@ai-sdk/openai-compatible` to Chat only. Catalog selection must become API-aware before Responses deployments can use it.
|
||||
3. Bedrock native auth is not AI SDK parity. The AI SDK plugin uses the default AWS provider chain, profile, container credentials, and Bedrock bearer token env behavior. Native Bedrock currently expects explicit credentials or bearer auth on the facade.
|
||||
4. Vertex Gemini, Vertex Chat, Vertex Responses, and Vertex Messages now have native package entrypoints, but the core runner does not map catalog metadata to them yet and recorded provider coverage is still missing.
|
||||
5. Azure is only a provider facade, not a full runtime replacement. Native Azure exists, but the catalog runner does not select it, and token auth/resource variants need review.
|
||||
6. Provider option typing is uneven. OpenAI, Anthropic, Gemini, Bedrock, and OpenRouter each expose a small typed subset plus raw HTTP overlays; this is useful but not equivalent to AI SDK provider option coverage.
|
||||
7. Structured output is not provider-native yet. `LLM.generateObject` still uses a synthetic tool strategy, while the future design expects native structured output where reliable and tool fallback where needed.
|
||||
8. Package/namespace boundaries for the current native loading set are explicit in docs and exports. Other exported provider facades are not catalog package entrypoints until they implement the contract. Vertex xAI still needs catalog API selection.
|
||||
9. Recorded coverage is uneven. OpenAI, Anthropic, Gemini, Bedrock Converse, Bedrock Mantle, Cloudflare, OpenRouter, and several OpenAI-compatible Chat providers have cassettes. Azure and Vertex still need first-class recorded scenarios before switching defaults.
|
||||
|
||||
## Native Namespace Shape
|
||||
|
||||
These are implementation/API slices, not separate npm packages.
|
||||
|
||||
| API slice | Package-like entrypoint | Purpose |
|
||||
| ----------------------------- | ----------------------------------------------------------- | ------------------------------------------------------------------------------------------ |
|
||||
| OpenAI Chat | `@opencode-ai/ai/providers/openai/chat` | OpenAI `/chat/completions` semantics. |
|
||||
| OpenAI Responses | `@opencode-ai/ai/providers/openai/responses` | OpenAI `/responses` semantics with HTTP default and optional per-call WebSocket execution. |
|
||||
| OpenAI-compatible Chat | `@opencode-ai/ai/providers/openai-compatible` | Generic OpenAI-compatible `/chat/completions`. |
|
||||
| Open Responses-compatible | `@opencode-ai/ai/providers/openai-compatible/responses` | Generic provider-neutral `/responses`. |
|
||||
| Anthropic-compatible Messages | `@opencode-ai/ai/providers/anthropic-compatible` | Generic Anthropic-compatible `/messages`. |
|
||||
| Anthropic Messages | `@opencode-ai/ai/providers/anthropic` | Anthropic Messages API. |
|
||||
| Gemini Developer API | `@opencode-ai/ai/providers/google` | Google AI Studio Gemini API. |
|
||||
| Vertex Gemini | `@opencode-ai/ai/providers/google-vertex/gemini` | Vertex Gemini API; `providers/google-vertex` is the default alias. |
|
||||
| Vertex Chat | `@opencode-ai/ai/providers/google-vertex/chat` | Vertex OpenAI-compatible Chat Completions for MaaS models. |
|
||||
| Vertex Responses | `@opencode-ai/ai/providers/google-vertex/responses` | Vertex Open Responses for Grok models. |
|
||||
| Vertex Messages | `@opencode-ai/ai/providers/google-vertex/messages` | Vertex-hosted Anthropic Messages API. |
|
||||
| Bedrock Converse | `@opencode-ai/ai/providers/amazon-bedrock` | AWS Bedrock Converse API. |
|
||||
| Bedrock Mantle Chat | `@opencode-ai/ai/providers/amazon-bedrock/mantle/chat` | AWS Bedrock Mantle OpenAI-compatible Chat API. |
|
||||
| Bedrock Mantle Responses | `@opencode-ai/ai/providers/amazon-bedrock/mantle/responses` | AWS Bedrock Mantle OpenAI-compatible Responses API. |
|
||||
| Azure OpenAI Chat | `@opencode-ai/ai/providers/azure/chat` | Azure specialization of OpenAI Chat. |
|
||||
| Azure OpenAI Responses | `@opencode-ai/ai/providers/azure/responses` | Azure specialization of OpenAI Responses. |
|
||||
|
||||
## Suggested Next Work Slices
|
||||
|
||||
1. Add native runner/catalog mappings for `@ai-sdk/azure`, `@ai-sdk/google`, and `@ai-sdk/amazon-bedrock` where the existing native facades are already close.
|
||||
2. Add API-aware runner/catalog selection between OpenAI-compatible Chat and Responses.
|
||||
3. Bring Bedrock native auth/config to AI SDK parity: region, profile, default AWS credential chain, bearer token env, endpoint override, and cross-region inference profile handling.
|
||||
4. Add runner/catalog mappings and recorded scenarios for the native Vertex Gemini, Chat, Responses, and Messages entrypoints.
|
||||
5. Decide Chat/Responses selection for `@ai-sdk/google-vertex/xai` catalog models.
|
||||
6. Expand typed provider options from the existing V1 lowerer knowledge in `packages/core/src/v1/config/provider-options.ts` before adding more raw overlay examples.
|
||||
7. Add recorded provider tests for Azure, Vertex Gemini, Vertex Chat, Vertex Responses, Vertex Messages, and Bedrock credential-chain behavior before making native runtime the default for those packages.
|
||||
@@ -1,38 +0,0 @@
|
||||
{
|
||||
"$schema": "https://json.schemastore.org/package.json",
|
||||
"version": "1.17.20",
|
||||
"name": "@opencode-ai/ai",
|
||||
"type": "module",
|
||||
"license": "MIT",
|
||||
"scripts": {
|
||||
"setup:recording-env": "bun run script/setup-recording-env.ts",
|
||||
"test": "bun test --timeout 30000 --only-failures",
|
||||
"typecheck": "tsgo --noEmit && tsgo --noEmit -p tsconfig.types.json",
|
||||
"build": "tsc -p tsconfig.build.json"
|
||||
},
|
||||
"files": [
|
||||
"dist"
|
||||
],
|
||||
"exports": {
|
||||
".": "./src/index.ts",
|
||||
"./testing": "./src/testing.ts",
|
||||
"./*": "./src/*.ts"
|
||||
},
|
||||
"devDependencies": {
|
||||
"@clack/prompts": "1.0.0-alpha.1",
|
||||
"@effect/platform-node": "catalog:",
|
||||
"@opencode-ai/http-recorder": "workspace:*",
|
||||
"@tsconfig/bun": "catalog:",
|
||||
"@types/bun": "catalog:",
|
||||
"@typescript/native-preview": "catalog:",
|
||||
"typescript": "catalog:"
|
||||
},
|
||||
"dependencies": {
|
||||
"@smithy/eventstream-codec": "4.2.14",
|
||||
"@smithy/util-utf8": "4.2.2",
|
||||
"@opencode-ai/schema": "workspace:*",
|
||||
"aws4fetch": "1.0.20",
|
||||
"effect": "catalog:",
|
||||
"google-auth-library": "10.5.0"
|
||||
}
|
||||
}
|
||||
@@ -1,38 +0,0 @@
|
||||
#!/usr/bin/env bun
|
||||
import { Script } from "@opencode-ai/script"
|
||||
import { $ } from "bun"
|
||||
import { fileURLToPath } from "url"
|
||||
|
||||
const dir = fileURLToPath(new URL("..", import.meta.url))
|
||||
process.chdir(dir)
|
||||
|
||||
async function published(name: string, version: string) {
|
||||
return (await $`npm view ${name}@${version} version`.nothrow()).exitCode === 0
|
||||
}
|
||||
|
||||
await $`bun run build`
|
||||
const originalText = await Bun.file("package.json").text()
|
||||
const pkg = JSON.parse(originalText) as {
|
||||
name: string
|
||||
version: string
|
||||
exports: Record<string, string>
|
||||
}
|
||||
if (await published(pkg.name, pkg.version)) {
|
||||
console.log(`already published ${pkg.name}@${pkg.version}`)
|
||||
} else {
|
||||
for (const [key, value] of Object.entries(pkg.exports)) {
|
||||
const file = value.replace("./src/", "./dist/").replace(".ts", "")
|
||||
// @ts-ignore
|
||||
pkg.exports[key] = {
|
||||
import: file + ".js",
|
||||
types: file + ".d.ts",
|
||||
}
|
||||
}
|
||||
await Bun.write("package.json", JSON.stringify(pkg, null, 2))
|
||||
try {
|
||||
await $`bun pm pack`
|
||||
await $`npm publish *.tgz --tag ${Script.channel} --access public`
|
||||
} finally {
|
||||
await Bun.write("package.json", originalText)
|
||||
}
|
||||
}
|
||||
@@ -1,149 +0,0 @@
|
||||
// Apply an `LLMRequest.cache` policy by injecting `CacheHint`s onto the parts
|
||||
// the policy designates. Runs once at compile time, before the per-protocol
|
||||
// body builder, so the existing inline-hint lowering path handles the rest.
|
||||
//
|
||||
// The default `"auto"` shape places breakpoints at the last tool definition,
|
||||
// the first and last distinct system parts, and the conversation tail. This
|
||||
// exposes reusable tool, base-agent, project, and session prefixes while
|
||||
// advancing the tail after each tool result keeps the previous cache entry
|
||||
// within Anthropic's 20-block lookback during long agent turns.
|
||||
//
|
||||
// Manual `cache: CacheHint` placements on individual parts are preserved and
|
||||
// count against the four-breakpoint budget; auto only fills remaining slots.
|
||||
import { CacheHint, type CachePolicy, type CachePolicyObject } from "./schema/options.js"
|
||||
import { LLMRequest, Message, ToolDefinition, type ContentPart } from "./schema/messages.js"
|
||||
|
||||
const AUTO: CachePolicyObject = {
|
||||
tools: true,
|
||||
system: true,
|
||||
messages: { tail: 1 },
|
||||
}
|
||||
|
||||
const NONE: CachePolicyObject = {}
|
||||
const BREAKPOINT_CAP = 4
|
||||
|
||||
// Resolution rules:
|
||||
// - undefined → "auto" — caching is on by default. The math favors it:
|
||||
// Anthropic 5m-cache write is 1.25x base, read is 0.1x,
|
||||
// so a single reuse within 5 minutes already wins.
|
||||
// - "auto" → tools + first/last system + final message boundary.
|
||||
// - "none" → no auto placement; manual `CacheHint`s still flow.
|
||||
// - object form → exactly what the caller asked for.
|
||||
const resolve = (policy: CachePolicy | undefined): CachePolicyObject => {
|
||||
if (policy === undefined || policy === "auto") return AUTO
|
||||
if (policy === "none") return NONE
|
||||
return policy
|
||||
}
|
||||
|
||||
// Protocols whose wire format ignores inline cache markers (OpenAI's implicit
|
||||
// prefix caching, Gemini's implicit + out-of-band CachedContent). Skip the
|
||||
// whole policy pass for these — emitting hints would be harmless but pointless.
|
||||
const RESPECTS_INLINE_HINTS = new Set(["anthropic-messages", "bedrock-converse", "openrouter"])
|
||||
|
||||
const makeHint = (ttlSeconds: number | undefined): CacheHint =>
|
||||
ttlSeconds !== undefined ? new CacheHint({ type: "ephemeral", ttlSeconds }) : new CacheHint({ type: "ephemeral" })
|
||||
|
||||
interface Budget {
|
||||
remaining: number
|
||||
}
|
||||
|
||||
const markLastTool = (
|
||||
tools: ReadonlyArray<ToolDefinition>,
|
||||
hint: CacheHint,
|
||||
budget: Budget,
|
||||
): ReadonlyArray<ToolDefinition> => {
|
||||
if (tools.length === 0) return tools
|
||||
const last = tools.length - 1
|
||||
if (tools[last]!.cache || budget.remaining === 0) return tools
|
||||
budget.remaining -= 1
|
||||
return tools.map((tool, i) => (i === last ? new ToolDefinition({ ...tool, cache: hint }) : tool))
|
||||
}
|
||||
|
||||
const markSystemBoundaries = (system: LLMRequest["system"], hint: CacheHint, budget: Budget): LLMRequest["system"] => {
|
||||
if (system.length === 0) return system
|
||||
let changed = false
|
||||
const next = system.map((part, index) => {
|
||||
if ((index !== 0 && index !== system.length - 1) || part.cache || budget.remaining === 0) return part
|
||||
budget.remaining -= 1
|
||||
changed = true
|
||||
return { ...part, cache: hint }
|
||||
})
|
||||
return changed ? next : system
|
||||
}
|
||||
|
||||
const lastIndexOfRole = (messages: ReadonlyArray<Message>, role: Message["role"]): number =>
|
||||
messages.findLastIndex((m) => m.role === role)
|
||||
|
||||
// Mark the last text part of `messages[index]`. If no text part exists, mark
|
||||
// the last content part regardless of type — that's the breakpoint position
|
||||
// in tool-result-only messages too.
|
||||
const markMessageAt = (
|
||||
messages: ReadonlyArray<Message>,
|
||||
index: number,
|
||||
hint: CacheHint,
|
||||
budget: Budget,
|
||||
): ReadonlyArray<Message> => {
|
||||
if (index < 0 || index >= messages.length) return messages
|
||||
const target = messages[index]!
|
||||
if (target.content.length === 0) return messages
|
||||
const lastTextIndex = target.content.findLastIndex((part) => part.type === "text")
|
||||
const markAt = lastTextIndex >= 0 ? lastTextIndex : target.content.length - 1
|
||||
const existing = target.content[markAt]!
|
||||
if (("cache" in existing && existing.cache) || budget.remaining === 0) return messages
|
||||
budget.remaining -= 1
|
||||
const nextContent = target.content.map((part, i) => (i === markAt ? ({ ...part, cache: hint } as ContentPart) : part))
|
||||
const next = new Message({ ...target, content: nextContent })
|
||||
// Single pass over `messages`, substituting the one updated entry. Long
|
||||
// conversations call this on every request, so avoid `.map()` here — its
|
||||
// closure dispatch and identity copies show up in profiling.
|
||||
const result = messages.slice()
|
||||
result[index] = next
|
||||
return result
|
||||
}
|
||||
|
||||
const markMessages = (
|
||||
messages: ReadonlyArray<Message>,
|
||||
strategy: NonNullable<CachePolicyObject["messages"]>,
|
||||
hint: CacheHint,
|
||||
budget: Budget,
|
||||
): ReadonlyArray<Message> => {
|
||||
if (messages.length === 0) return messages
|
||||
if (strategy === "latest-user-message")
|
||||
return markMessageAt(messages, lastIndexOfRole(messages, "user"), hint, budget)
|
||||
if (strategy === "latest-assistant")
|
||||
return markMessageAt(messages, lastIndexOfRole(messages, "assistant"), hint, budget)
|
||||
const start = Math.max(0, messages.length - strategy.tail)
|
||||
let next = messages
|
||||
for (let i = start; i < messages.length; i++) next = markMessageAt(next, i, hint, budget)
|
||||
return next
|
||||
}
|
||||
|
||||
const countHints = (request: LLMRequest) =>
|
||||
request.tools.reduce((count, tool) => count + (tool.cache === undefined ? 0 : 1), 0) +
|
||||
request.system.reduce((count, part) => count + (part.cache === undefined ? 0 : 1), 0) +
|
||||
request.messages.reduce(
|
||||
(count, message) =>
|
||||
count +
|
||||
message.content.reduce(
|
||||
(contentCount, part) => contentCount + ("cache" in part && part.cache !== undefined ? 1 : 0),
|
||||
0,
|
||||
),
|
||||
0,
|
||||
)
|
||||
|
||||
export const applyCachePolicy = (request: LLMRequest): LLMRequest => {
|
||||
if (!RESPECTS_INLINE_HINTS.has(request.model.route.id)) return request
|
||||
if (request.model.route.id === "openrouter" && (request.cache === undefined || request.cache === "auto"))
|
||||
return request
|
||||
const policy = resolve(request.cache)
|
||||
if (!policy.tools && !policy.system && !policy.messages) return request
|
||||
|
||||
const hint = makeHint(policy.ttlSeconds)
|
||||
const budget = { remaining: Math.max(0, BREAKPOINT_CAP - countHints(request)) }
|
||||
const tools = policy.tools ? markLastTool(request.tools, hint, budget) : request.tools
|
||||
const system = policy.system ? markSystemBoundaries(request.system, hint, budget) : request.system
|
||||
const messages = policy.messages ? markMessages(request.messages, policy.messages, hint, budget) : request.messages
|
||||
|
||||
if (tools === request.tools && system === request.system && messages === request.messages) return request
|
||||
return LLMRequest.update(request, { tools, system, messages })
|
||||
}
|
||||
@@ -1,38 +0,0 @@
|
||||
import { Context, Effect, Layer } from "effect"
|
||||
import { RequestExecutor } from "./route/executor.js"
|
||||
import type { ImageOptions, ImageRequest, ImageRequestFor, ImageResponse } from "./image.js"
|
||||
import type { AIError } from "./schema/index.js"
|
||||
|
||||
export type Execute = RequestExecutor.Interface["execute"]
|
||||
|
||||
export interface Interface {
|
||||
readonly generate: <Options extends ImageOptions>(
|
||||
request: ImageRequestFor<Options>,
|
||||
) => Effect.Effect<ImageResponse, AIError>
|
||||
}
|
||||
|
||||
export class Service extends Context.Service<Service, Interface>()("@opencode/ImageClient") {}
|
||||
|
||||
export const generate = <Options extends ImageOptions>(
|
||||
request: ImageRequestFor<Options>,
|
||||
): Effect.Effect<ImageResponse, AIError, Service> =>
|
||||
Effect.gen(function* () {
|
||||
const client = yield* Service
|
||||
return yield* client.generate(request)
|
||||
})
|
||||
|
||||
export const layer: Layer.Layer<Service, never, RequestExecutor.Service> = Layer.effect(
|
||||
Service,
|
||||
Effect.gen(function* () {
|
||||
const executor = yield* RequestExecutor.Service
|
||||
return Service.of({
|
||||
generate: (request) => request.model.route.generate(request, executor.execute),
|
||||
})
|
||||
}),
|
||||
)
|
||||
|
||||
export const ImageClient = {
|
||||
Service,
|
||||
layer,
|
||||
generate,
|
||||
} as const
|
||||
@@ -1,171 +0,0 @@
|
||||
import { Effect, Schema } from "effect"
|
||||
import {
|
||||
HttpOptions,
|
||||
InvalidRequestReason,
|
||||
AIError,
|
||||
ModelID,
|
||||
ProviderID,
|
||||
ProviderMetadata,
|
||||
Usage,
|
||||
} from "./schema/index.js"
|
||||
import { ImageClient, Service, type Execute as ImageExecute } from "./image-client.js"
|
||||
|
||||
export interface ImageRoute<Options extends ImageOptions = ImageOptions> {
|
||||
readonly id: string
|
||||
readonly generate: (request: ImageRequestFor<Options>, execute: ImageExecute) => Effect.Effect<ImageResponse, AIError>
|
||||
}
|
||||
|
||||
export type ImageOptions = Record<string, unknown>
|
||||
|
||||
export class ImageModel<Options extends ImageOptions = ImageOptions> {
|
||||
declare protected readonly _Options: (options: Options) => Options
|
||||
readonly id: ModelID
|
||||
readonly provider: ProviderID
|
||||
readonly route: ImageRoute<Options>
|
||||
readonly http?: HttpOptions
|
||||
|
||||
constructor(input: ImageModel.Input<Options>) {
|
||||
this.id = input.id
|
||||
this.provider = input.provider
|
||||
this.route = input.route
|
||||
this.http = input.http
|
||||
}
|
||||
|
||||
static make<Options extends ImageOptions = ImageOptions>(input: ImageModel.MakeInput<Options>) {
|
||||
return new ImageModel<Options>({
|
||||
id: ModelID.make(input.id),
|
||||
provider: ProviderID.make(input.provider),
|
||||
route: input.route,
|
||||
http: input.http,
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
export namespace ImageModel {
|
||||
export interface Input<Options extends ImageOptions = ImageOptions> {
|
||||
readonly id: ModelID
|
||||
readonly provider: ProviderID
|
||||
readonly route: ImageRoute<Options>
|
||||
readonly http?: HttpOptions
|
||||
}
|
||||
|
||||
export interface MakeInput<Options extends ImageOptions = ImageOptions>
|
||||
extends Omit<Input<Options>, "id" | "provider"> {
|
||||
readonly id: string | ModelID
|
||||
readonly provider: string | ProviderID
|
||||
}
|
||||
}
|
||||
|
||||
export const ImageModelSchema = Schema.declare((value): value is ImageModel => value instanceof ImageModel, {
|
||||
expected: "Image.Model",
|
||||
})
|
||||
|
||||
const ImageBytesInput = Schema.Struct({
|
||||
type: Schema.Literal("bytes"),
|
||||
data: Schema.Uint8Array,
|
||||
mediaType: Schema.String,
|
||||
})
|
||||
const ImageUrlInput = Schema.Struct({
|
||||
type: Schema.Literal("url"),
|
||||
url: Schema.String,
|
||||
})
|
||||
const ImageFileIDInput = Schema.Struct({
|
||||
type: Schema.Literal("file-id"),
|
||||
id: Schema.String,
|
||||
})
|
||||
const ImageFileURIInput = Schema.Struct({
|
||||
type: Schema.Literal("file-uri"),
|
||||
uri: Schema.String,
|
||||
mediaType: Schema.String,
|
||||
})
|
||||
|
||||
export const ImageInputSchema = Schema.Union([
|
||||
ImageBytesInput,
|
||||
ImageUrlInput,
|
||||
ImageFileIDInput,
|
||||
ImageFileURIInput,
|
||||
]).pipe(Schema.toTaggedUnion("type"))
|
||||
export type ImageInput = Schema.Schema.Type<typeof ImageInputSchema>
|
||||
|
||||
export const ImageInput = {
|
||||
bytes: (data: Uint8Array, mediaType: string): ImageInput => ({ type: "bytes", data, mediaType }),
|
||||
url: (url: string): ImageInput => ({ type: "url", url }),
|
||||
file: (id: string): ImageInput => ({ type: "file-id", id }),
|
||||
fileUri: (uri: string, mediaType: string): ImageInput => ({ type: "file-uri", uri, mediaType }),
|
||||
} as const
|
||||
|
||||
export class ImageRequest extends Schema.Class<ImageRequest>("Image.Request")({
|
||||
model: ImageModelSchema,
|
||||
prompt: Schema.String,
|
||||
images: Schema.optional(Schema.Array(ImageInputSchema)),
|
||||
options: Schema.optional(Schema.Record(Schema.String, Schema.Unknown)),
|
||||
http: Schema.optional(HttpOptions),
|
||||
}) {
|
||||
declare protected readonly _ImageRequest: void
|
||||
}
|
||||
|
||||
export type ImageRequestFor<Options extends ImageOptions = ImageOptions> = Omit<ImageRequest, "model" | "options"> & {
|
||||
readonly model: ImageModel<Options>
|
||||
readonly options?: Options
|
||||
}
|
||||
|
||||
export type ImageModelOptions<Model> = Model extends ImageModel<infer Options> ? Options : never
|
||||
|
||||
export type ImageRequestInput<Model extends object = ImageModel> = Omit<
|
||||
ConstructorParameters<typeof ImageRequest>[0],
|
||||
"model" | "options" | "http"
|
||||
> & {
|
||||
readonly model: Model
|
||||
readonly options?: NoInfer<ImageModelOptions<Model>>
|
||||
readonly http?: HttpOptions.Input
|
||||
} & (Model extends ImageModel<ImageModelOptions<Model>> ? unknown : never)
|
||||
|
||||
export class GeneratedImage extends Schema.Class<GeneratedImage>("Image.Generated")({
|
||||
mediaType: Schema.String,
|
||||
data: Schema.Union([Schema.String, Schema.Uint8Array]),
|
||||
providerMetadata: Schema.optional(ProviderMetadata),
|
||||
}) {}
|
||||
|
||||
export class ImageResponse extends Schema.Class<ImageResponse>("Image.Response")({
|
||||
images: Schema.Array(GeneratedImage),
|
||||
usage: Schema.optional(Usage),
|
||||
providerMetadata: Schema.optional(ProviderMetadata),
|
||||
}) {
|
||||
get image() {
|
||||
return this.images[0]
|
||||
}
|
||||
}
|
||||
|
||||
export function request<const Model extends object>(
|
||||
input: ImageRequestInput<Model>,
|
||||
): ImageRequestFor<ImageModelOptions<Model>>
|
||||
export function request(input: ImageRequest): ImageRequest
|
||||
export function request(input: ImageRequest | ImageRequestInput) {
|
||||
if (input instanceof ImageRequest) return input
|
||||
return new ImageRequest({
|
||||
...input,
|
||||
model: input.model as unknown as ImageModel,
|
||||
http: input.http === undefined ? undefined : HttpOptions.make(input.http),
|
||||
})
|
||||
}
|
||||
|
||||
export function generate<const Model extends object>(
|
||||
input: ImageRequestInput<Model>,
|
||||
): Effect.Effect<ImageResponse, AIError, Service>
|
||||
export function generate(input: ImageRequest): Effect.Effect<ImageResponse, AIError, Service>
|
||||
export function generate(input: ImageRequest | ImageRequestInput) {
|
||||
return Effect.try({
|
||||
try: () => (input instanceof ImageRequest ? input : request(input)),
|
||||
catch: (error) =>
|
||||
new AIError({
|
||||
module: "Image",
|
||||
method: "generate",
|
||||
reason: new InvalidRequestReason({ message: error instanceof Error ? error.message : String(error) }),
|
||||
}),
|
||||
}).pipe(Effect.flatMap((request) => ImageClient.generate(request as unknown as ImageRequestFor<ImageOptions>)))
|
||||
}
|
||||
|
||||
export const Image = {
|
||||
request,
|
||||
generate,
|
||||
} as const
|
||||
@@ -1,42 +0,0 @@
|
||||
export { LLMClient } from "./route/client.js"
|
||||
export { ImageClient } from "./image-client.js"
|
||||
export { Auth } from "./route/auth.js"
|
||||
export { Provider } from "./provider.js"
|
||||
export { ProviderPackage } from "./provider-package.js"
|
||||
export { isContextOverflow, isContextOverflowFailure } from "./provider-error.js"
|
||||
export type {
|
||||
RouteLanguageModelInput,
|
||||
RouteRoutedLanguageModelInput,
|
||||
Interface as LLMClientShape,
|
||||
Service as LLMClientService,
|
||||
} from "./route/client.js"
|
||||
export * from "./schema/index.js"
|
||||
export { GeneratedImage, ImageInput, ImageInputSchema, ImageModel, ImageRequest, ImageResponse } from "./image.js"
|
||||
export type { ImageModelOptions, ImageOptions, ImageRequestFor, ImageRequestInput, ImageRoute } from "./image.js"
|
||||
export { Image } from "./image.js"
|
||||
export { Tool, ToolFailure, toDefinitions } from "./tool.js"
|
||||
export { ToolRuntime } from "./tool-runtime.js"
|
||||
export type { DispatchResult as ToolDispatchResult, ToolSettlement } from "./tool-runtime.js"
|
||||
export type {
|
||||
AnyExecutableTool,
|
||||
AnyTool,
|
||||
ExecutableTool,
|
||||
ExecutableTools,
|
||||
Definition as ToolShape,
|
||||
ToolExecute,
|
||||
ToolExecuteContext,
|
||||
ToolModelOutputInput,
|
||||
Tools,
|
||||
ToolSchema,
|
||||
ToolToModelOutput,
|
||||
} from "./tool.js"
|
||||
export * as LLM from "./llm.js"
|
||||
export type {
|
||||
Definition as ProviderDefinition,
|
||||
LanguageModelFactory as ProviderLanguageModelFactory,
|
||||
LanguageModelOptions as ProviderLanguageModelOptions,
|
||||
} from "./provider.js"
|
||||
export type {
|
||||
Definition as ProviderPackageDefinition,
|
||||
Settings as ProviderPackageSettings,
|
||||
} from "./provider-package.js"
|
||||
@@ -1,180 +0,0 @@
|
||||
import { Effect, JsonSchema, Schema } from "effect"
|
||||
import { LLMClient, Service } from "./route/client.js"
|
||||
import {
|
||||
GenerationOptions,
|
||||
HttpOptions,
|
||||
InvalidProviderOutputReason,
|
||||
AIError,
|
||||
LLMEvent,
|
||||
LLMRequest,
|
||||
LLMResponse,
|
||||
Message,
|
||||
LanguageModel,
|
||||
SystemPart,
|
||||
ToolChoice,
|
||||
ToolDefinition,
|
||||
type ContentPart,
|
||||
type LanguageModelProviderOptions,
|
||||
} from "./schema/index.js"
|
||||
import { make as makeTool, toDefinitions, type ToolSchema } from "./tool.js"
|
||||
|
||||
/** Input accepted by `LLM.request`, normalized into the canonical `LLMRequest` class. */
|
||||
export type RequestInput<SelectedLanguageModel extends LanguageModel = LanguageModel> = Omit<
|
||||
ConstructorParameters<typeof LLMRequest>[0],
|
||||
"model" | "system" | "messages" | "tools" | "toolChoice" | "generation" | "http" | "providerOptions"
|
||||
> & {
|
||||
readonly model: SelectedLanguageModel
|
||||
readonly system?: string | SystemPart | ReadonlyArray<SystemPart>
|
||||
readonly prompt?: string | ContentPart | ReadonlyArray<ContentPart>
|
||||
readonly messages?: ReadonlyArray<Message | Message.Input>
|
||||
readonly tools?: ReadonlyArray<ToolDefinition.Input>
|
||||
readonly toolChoice?: ToolChoice.Input
|
||||
readonly generation?: GenerationOptions.Input
|
||||
readonly providerOptions?: NoInfer<LanguageModelProviderOptions<SelectedLanguageModel>>
|
||||
readonly http?: HttpOptions.Input
|
||||
}
|
||||
|
||||
export const generate = LLMClient.generate
|
||||
|
||||
export const stream = LLMClient.stream
|
||||
|
||||
export const request = <const SelectedLanguageModel extends LanguageModel>(
|
||||
input: RequestInput<SelectedLanguageModel>,
|
||||
) => {
|
||||
const {
|
||||
system: requestSystem,
|
||||
prompt,
|
||||
messages,
|
||||
tools,
|
||||
toolChoice: requestToolChoice,
|
||||
generation: requestGeneration,
|
||||
providerOptions: requestProviderOptions,
|
||||
http: requestHttp,
|
||||
...rest
|
||||
} = input
|
||||
return new LLMRequest({
|
||||
...rest,
|
||||
system: SystemPart.content(requestSystem),
|
||||
messages: [...(messages?.map(Message.make) ?? []), ...(prompt === undefined ? [] : [Message.user(prompt)])],
|
||||
tools: tools?.map(ToolDefinition.make) ?? [],
|
||||
toolChoice: requestToolChoice ? ToolChoice.make(requestToolChoice) : undefined,
|
||||
generation: requestGeneration === undefined ? undefined : GenerationOptions.make(requestGeneration),
|
||||
providerOptions: requestProviderOptions,
|
||||
http: requestHttp === undefined ? undefined : HttpOptions.make(requestHttp),
|
||||
})
|
||||
}
|
||||
|
||||
const GENERATE_OBJECT_TOOL_NAME = "generate_object"
|
||||
|
||||
const GENERATE_OBJECT_TOOL_DESCRIPTION = "Return the structured result by calling this tool."
|
||||
|
||||
type GenerateObjectBase<SelectedLanguageModel extends LanguageModel = LanguageModel> = Omit<
|
||||
RequestInput<SelectedLanguageModel>,
|
||||
"tools" | "toolChoice"
|
||||
>
|
||||
|
||||
export class GenerateObjectResponse<T> {
|
||||
constructor(
|
||||
readonly object: T,
|
||||
readonly response: LLMResponse,
|
||||
) {}
|
||||
|
||||
get events() {
|
||||
return this.response.events
|
||||
}
|
||||
|
||||
get usage() {
|
||||
return this.response.usage
|
||||
}
|
||||
}
|
||||
|
||||
export interface GenerateObjectOptions<
|
||||
S extends ToolSchema<any>,
|
||||
SelectedLanguageModel extends LanguageModel = LanguageModel,
|
||||
> extends GenerateObjectBase<SelectedLanguageModel> {
|
||||
readonly schema: S
|
||||
}
|
||||
|
||||
export interface GenerateObjectDynamicOptions<SelectedLanguageModel extends LanguageModel = LanguageModel>
|
||||
extends GenerateObjectBase<SelectedLanguageModel> {
|
||||
/** Raw JSON Schema object describing the expected output shape. */
|
||||
readonly jsonSchema: JsonSchema.JsonSchema
|
||||
}
|
||||
|
||||
const runGenerateObject = Effect.fn("LLM.generateObject")(function* (
|
||||
options: GenerateObjectBase,
|
||||
tool: ReturnType<typeof makeTool>,
|
||||
) {
|
||||
const baseRequest = request(options)
|
||||
const generateRequest = LLMRequest.update(baseRequest, {
|
||||
tools: toDefinitions({ [GENERATE_OBJECT_TOOL_NAME]: tool }),
|
||||
toolChoice: ToolChoice.named(GENERATE_OBJECT_TOOL_NAME),
|
||||
})
|
||||
const response = yield* LLMClient.generate(generateRequest)
|
||||
const call = response.toolCalls.find(
|
||||
(event) => LLMEvent.is.toolCall(event) && event.name === GENERATE_OBJECT_TOOL_NAME,
|
||||
)
|
||||
if (!call || !LLMEvent.is.toolCall(call))
|
||||
return yield* new AIError({
|
||||
module: "LLM",
|
||||
method: "generateObject",
|
||||
reason: new InvalidProviderOutputReason({
|
||||
message: `generateObject: model did not call the forced \`${GENERATE_OBJECT_TOOL_NAME}\` tool`,
|
||||
}),
|
||||
})
|
||||
const object = yield* tool._decode(call.input).pipe(
|
||||
Effect.mapError(
|
||||
(error) =>
|
||||
new AIError({
|
||||
module: "LLM",
|
||||
method: "generateObject",
|
||||
reason: new InvalidProviderOutputReason({
|
||||
message: `generateObject: tool input failed schema decode: ${error.message}`,
|
||||
}),
|
||||
}),
|
||||
),
|
||||
)
|
||||
return new GenerateObjectResponse(object, response)
|
||||
})
|
||||
|
||||
/**
|
||||
* Run a model and decode its output against `schema`. Works on every protocol
|
||||
* because it forces a synthetic tool call internally — provider-native JSON
|
||||
* modes are intentionally avoided so behaviour is uniform.
|
||||
*
|
||||
* Two input modes:
|
||||
*
|
||||
* 1. `schema: EffectSchema<T>` — `.object` is decoded and typed as `T`.
|
||||
* Decode failures surface as `AIError`.
|
||||
* 2. `jsonSchema: JsonSchema.JsonSchema` — `.object` is `unknown`. Use when
|
||||
* the schema is only available at runtime (MCP, plugin manifests). Caller validates.
|
||||
*/
|
||||
export function generateObject<const SelectedLanguageModel extends LanguageModel, S extends ToolSchema<any>>(
|
||||
options: GenerateObjectOptions<S, SelectedLanguageModel>,
|
||||
): Effect.Effect<GenerateObjectResponse<Schema.Schema.Type<S>>, AIError, Service>
|
||||
export function generateObject<const SelectedLanguageModel extends LanguageModel>(
|
||||
options: GenerateObjectDynamicOptions<SelectedLanguageModel>,
|
||||
): Effect.Effect<GenerateObjectResponse<unknown>, AIError, Service>
|
||||
export function generateObject(options: GenerateObjectOptions<ToolSchema<any>> | GenerateObjectDynamicOptions) {
|
||||
if ("schema" in options) {
|
||||
const { schema, ...rest } = options
|
||||
return runGenerateObject(
|
||||
rest,
|
||||
makeTool({
|
||||
description: GENERATE_OBJECT_TOOL_DESCRIPTION,
|
||||
parameters: schema,
|
||||
success: Schema.Unknown as ToolSchema<unknown>,
|
||||
execute: () => Effect.void,
|
||||
}),
|
||||
)
|
||||
}
|
||||
const { jsonSchema, ...rest } = options
|
||||
return runGenerateObject(
|
||||
rest,
|
||||
makeTool({
|
||||
description: GENERATE_OBJECT_TOOL_DESCRIPTION,
|
||||
jsonSchema,
|
||||
execute: () => Effect.void,
|
||||
}),
|
||||
)
|
||||
}
|
||||
@@ -1 +0,0 @@
|
||||
export * from "./protocols/index.js"
|
||||
File diff suppressed because it is too large
Load Diff
@@ -1,646 +0,0 @@
|
||||
import { Effect, Schema } from "effect"
|
||||
import { Tool } from "@opencode-ai/schema/tool"
|
||||
import { Route } from "../route/client.js"
|
||||
import { Auth } from "../route/auth.js"
|
||||
import { Endpoint } from "../route/endpoint.js"
|
||||
import { Framing } from "../route/framing.js"
|
||||
import { Protocol } from "../route/protocol.js"
|
||||
import {
|
||||
LLMEvent,
|
||||
Usage,
|
||||
type FinishReason,
|
||||
type JsonSchema,
|
||||
type LLMRequest,
|
||||
type MediaPart,
|
||||
type ProviderOptions,
|
||||
type ProviderMetadata,
|
||||
type TextPart,
|
||||
type ToolCallPart,
|
||||
type ToolDefinition,
|
||||
} from "../schema/index.js"
|
||||
import { JsonObject, optionalArray, ProviderShared } from "./shared.js"
|
||||
import { GeminiToolSchema } from "./utils/gemini-tool-schema.js"
|
||||
import { Lifecycle } from "./utils/lifecycle.js"
|
||||
import { ToolSchemaProjection } from "./utils/tool-schema.js"
|
||||
|
||||
const ADAPTER = "gemini"
|
||||
const MEDIA_MIMES = new Set<string>(ProviderShared.MEDIA_MIMES)
|
||||
// Google documents this sentinel for replaying Gemini 3 function calls after their original signature was lost.
|
||||
const SKIP_THOUGHT_SIGNATURE_VALIDATOR = "skip_thought_signature_validator"
|
||||
export const DEFAULT_BASE_URL = "https://generativelanguage.googleapis.com/v1beta"
|
||||
|
||||
// Gemini 3 rejects replayed function calls without a thought signature. Google's SDKs avoid that in normal chats by
|
||||
// retaining complete model responses, but OpenCode reconstructs durable history and may encounter an unsigned call
|
||||
// from an older or external session. Model IDs are open-ended, so unknown Gemini aliases inherit current behavior.
|
||||
const requiresThoughtSignatureFallback = (modelID: string) => {
|
||||
if (!/(^|\/)gemini-/i.test(modelID)) return false
|
||||
if (/(^|\/)gemini-(?:1|2)(?:[.-]|$)/i.test(modelID)) return false
|
||||
if (/(^|\/)gemini-pro(?:-vision)?$/i.test(modelID)) return false
|
||||
return !/(^|\/)gemini-robotics-er-1\.5(?:[.-]|$)/i.test(modelID)
|
||||
}
|
||||
|
||||
export interface OptionsInput {
|
||||
readonly [key: string]: unknown
|
||||
readonly cachedContent?: string
|
||||
readonly safetySettings?: ReadonlyArray<{
|
||||
readonly category:
|
||||
| "HARM_CATEGORY_UNSPECIFIED"
|
||||
| "HARM_CATEGORY_HATE_SPEECH"
|
||||
| "HARM_CATEGORY_DANGEROUS_CONTENT"
|
||||
| "HARM_CATEGORY_HARASSMENT"
|
||||
| "HARM_CATEGORY_SEXUALLY_EXPLICIT"
|
||||
| "HARM_CATEGORY_CIVIC_INTEGRITY"
|
||||
| (string & {})
|
||||
readonly threshold:
|
||||
| "HARM_BLOCK_THRESHOLD_UNSPECIFIED"
|
||||
| "BLOCK_LOW_AND_ABOVE"
|
||||
| "BLOCK_MEDIUM_AND_ABOVE"
|
||||
| "BLOCK_ONLY_HIGH"
|
||||
| "BLOCK_NONE"
|
||||
| "OFF"
|
||||
| (string & {})
|
||||
}>
|
||||
readonly serviceTier?: "standard" | "flex" | "priority" | (string & {})
|
||||
readonly thinkingConfig?: {
|
||||
readonly thinkingBudget?: number
|
||||
readonly includeThoughts?: boolean
|
||||
readonly thinkingLevel?: "minimal" | "low" | "medium" | "high" | (string & {})
|
||||
}
|
||||
}
|
||||
|
||||
export type ProviderOptionsInput = ProviderOptions & {
|
||||
readonly gemini?: OptionsInput
|
||||
}
|
||||
|
||||
// =============================================================================
|
||||
// Request Body Schema
|
||||
// =============================================================================
|
||||
const GeminiTextPart = Schema.Struct({
|
||||
text: Schema.String,
|
||||
thought: Schema.optional(Schema.Boolean),
|
||||
thoughtSignature: Schema.optional(Schema.String),
|
||||
})
|
||||
|
||||
const GeminiInlineDataPart = Schema.Struct({
|
||||
inlineData: Schema.Struct({
|
||||
mimeType: Schema.String,
|
||||
data: Schema.String,
|
||||
}),
|
||||
})
|
||||
type GeminiInlineDataPart = Schema.Schema.Type<typeof GeminiInlineDataPart>
|
||||
|
||||
const GeminiFunctionCallPart = Schema.Struct({
|
||||
functionCall: Schema.Struct({
|
||||
id: Schema.optional(Schema.String),
|
||||
name: Schema.String,
|
||||
args: Schema.Unknown,
|
||||
}),
|
||||
thoughtSignature: Schema.optional(Schema.String),
|
||||
})
|
||||
|
||||
const GeminiFunctionResponsePart = Schema.Struct({
|
||||
functionResponse: Schema.Struct({
|
||||
id: Schema.optional(Schema.String),
|
||||
name: Schema.String,
|
||||
response: Schema.Unknown,
|
||||
parts: Schema.optional(Schema.Array(GeminiInlineDataPart)),
|
||||
}),
|
||||
})
|
||||
|
||||
const GeminiContentPart = Schema.Union([
|
||||
GeminiTextPart,
|
||||
GeminiInlineDataPart,
|
||||
GeminiFunctionCallPart,
|
||||
GeminiFunctionResponsePart,
|
||||
])
|
||||
|
||||
const GeminiContent = Schema.Struct({
|
||||
role: Schema.Literals(["user", "model"]),
|
||||
parts: Schema.Array(GeminiContentPart),
|
||||
})
|
||||
type GeminiContent = Schema.Schema.Type<typeof GeminiContent>
|
||||
|
||||
const GeminiSystemInstruction = Schema.Struct({
|
||||
parts: Schema.Array(Schema.Struct({ text: Schema.String })),
|
||||
})
|
||||
|
||||
const GeminiFunctionDeclaration = Schema.Struct({
|
||||
name: Schema.String,
|
||||
description: Schema.String,
|
||||
parameters: Schema.optional(JsonObject),
|
||||
})
|
||||
|
||||
const GeminiTool = Schema.Struct({
|
||||
functionDeclarations: Schema.Array(GeminiFunctionDeclaration),
|
||||
})
|
||||
|
||||
const GeminiToolConfig = Schema.Struct({
|
||||
functionCallingConfig: Schema.Struct({
|
||||
mode: Schema.Literals(["AUTO", "NONE", "ANY"]),
|
||||
allowedFunctionNames: optionalArray(Schema.String),
|
||||
}),
|
||||
})
|
||||
|
||||
const GeminiThinkingConfig = Schema.Struct({
|
||||
thinkingBudget: Schema.optional(Schema.Number),
|
||||
includeThoughts: Schema.optional(Schema.Boolean),
|
||||
thinkingLevel: Schema.optional(Schema.String),
|
||||
})
|
||||
|
||||
const GeminiSafetySetting = Schema.Struct({
|
||||
category: Schema.String,
|
||||
threshold: Schema.String,
|
||||
})
|
||||
|
||||
const GeminiGenerationConfig = Schema.Struct({
|
||||
maxOutputTokens: Schema.optional(Schema.Number),
|
||||
temperature: Schema.optional(Schema.Number),
|
||||
topP: Schema.optional(Schema.Number),
|
||||
topK: Schema.optional(Schema.Number),
|
||||
frequencyPenalty: Schema.optional(Schema.Number),
|
||||
presencePenalty: Schema.optional(Schema.Number),
|
||||
seed: Schema.optional(Schema.Number),
|
||||
stopSequences: optionalArray(Schema.String),
|
||||
thinkingConfig: Schema.optional(GeminiThinkingConfig),
|
||||
})
|
||||
|
||||
const GeminiBodyFields = {
|
||||
cachedContent: Schema.optional(Schema.String),
|
||||
contents: Schema.Array(GeminiContent),
|
||||
safetySettings: optionalArray(GeminiSafetySetting),
|
||||
serviceTier: Schema.optional(Schema.String),
|
||||
systemInstruction: Schema.optional(GeminiSystemInstruction),
|
||||
tools: optionalArray(GeminiTool),
|
||||
toolConfig: Schema.optional(GeminiToolConfig),
|
||||
generationConfig: Schema.optional(GeminiGenerationConfig),
|
||||
}
|
||||
const GeminiBody = Schema.Struct(GeminiBodyFields)
|
||||
export type GeminiBody = Schema.Schema.Type<typeof GeminiBody>
|
||||
|
||||
const GeminiUsage = Schema.Struct({
|
||||
cachedContentTokenCount: Schema.optional(Schema.Number),
|
||||
thoughtsTokenCount: Schema.optional(Schema.Number),
|
||||
promptTokenCount: Schema.optional(Schema.Number),
|
||||
candidatesTokenCount: Schema.optional(Schema.Number),
|
||||
totalTokenCount: Schema.optional(Schema.Number),
|
||||
})
|
||||
type GeminiUsage = Schema.Schema.Type<typeof GeminiUsage>
|
||||
|
||||
const GeminiCandidate = Schema.Struct({
|
||||
content: Schema.optional(GeminiContent),
|
||||
finishReason: Schema.optional(Schema.String),
|
||||
})
|
||||
|
||||
const GeminiEvent = Schema.Struct({
|
||||
candidates: optionalArray(GeminiCandidate),
|
||||
usageMetadata: Schema.optional(GeminiUsage),
|
||||
})
|
||||
type GeminiEvent = Schema.Schema.Type<typeof GeminiEvent>
|
||||
|
||||
interface ParserState {
|
||||
readonly finishReason?: string
|
||||
readonly hasToolCalls: boolean
|
||||
readonly nextToolCallId: number
|
||||
readonly usage?: Usage
|
||||
readonly lifecycle: Lifecycle.State
|
||||
readonly reasoningSignature?: string
|
||||
}
|
||||
|
||||
// =============================================================================
|
||||
// Tool Schema Conversion
|
||||
// =============================================================================
|
||||
// Tool-schema conversion has two distinct concerns:
|
||||
//
|
||||
// 1. Sanitize — fix common authoring mistakes Gemini rejects: integer/number
|
||||
// enums (must be strings), `required` entries that don't match a property,
|
||||
// untyped arrays (`items` must be present), and `properties`/`required`
|
||||
// keys on non-object scalars. Mirrors OpenCode's historical Gemini rules.
|
||||
//
|
||||
// 2. Project — lossy mapping from JSON Schema to Gemini's schema dialect:
|
||||
// drop empty root parameter schemas while preserving nested empty objects,
|
||||
// expand type arrays into `anyOf`, derive `nullable: true` from null members,
|
||||
// coerce `const` to `[const]` enum, recurse properties/items, and propagate
|
||||
// only an allowlisted set of keys (description, required, format, type,
|
||||
// nullable, enum, properties, items, allOf, anyOf, oneOf, minLength).
|
||||
// Anything outside the allowlist (e.g. `additionalProperties`, `$ref`) is
|
||||
// silently dropped.
|
||||
//
|
||||
// Sanitize runs first, then project. The implementation lives in
|
||||
// `utils/gemini-tool-schema` so this protocol keeps the same shape as the other
|
||||
// provider protocols.
|
||||
|
||||
// =============================================================================
|
||||
// Request Lowering
|
||||
// =============================================================================
|
||||
const lowerTool = (tool: ToolDefinition, inputSchema: JsonSchema) => ({
|
||||
name: tool.name,
|
||||
description: tool.description,
|
||||
parameters: GeminiToolSchema.convert(inputSchema),
|
||||
})
|
||||
|
||||
const lowerToolConfig = (toolChoice: NonNullable<LLMRequest["toolChoice"]>) =>
|
||||
ProviderShared.matchToolChoice("Gemini", toolChoice, {
|
||||
auto: () => ({ functionCallingConfig: { mode: "AUTO" as const } }),
|
||||
none: () => ({ functionCallingConfig: { mode: "NONE" as const } }),
|
||||
required: () => ({ functionCallingConfig: { mode: "ANY" as const } }),
|
||||
tool: (name) => ({ functionCallingConfig: { mode: "ANY" as const, allowedFunctionNames: [name] } }),
|
||||
})
|
||||
|
||||
const lowerUserPart = Effect.fn("Gemini.lowerUserPart")(function* (part: TextPart | MediaPart) {
|
||||
if (part.type === "text") return { text: part.text }
|
||||
const media = yield* ProviderShared.validateMedia("Gemini", part, MEDIA_MIMES)
|
||||
return { inlineData: { mimeType: media.mime, data: media.base64 } }
|
||||
})
|
||||
|
||||
const googleMetadata = (metadata: Record<string, unknown>): ProviderMetadata => ({ google: metadata })
|
||||
|
||||
const thoughtSignature = (providerMetadata: ProviderMetadata | undefined) => {
|
||||
const google = providerMetadata?.google
|
||||
return ProviderShared.isRecord(google) && typeof google.thoughtSignature === "string"
|
||||
? google.thoughtSignature
|
||||
: undefined
|
||||
}
|
||||
|
||||
const functionCallId = (providerMetadata: ProviderMetadata | undefined) => {
|
||||
const google = providerMetadata?.google
|
||||
return ProviderShared.isRecord(google) && typeof google.functionCallId === "string"
|
||||
? google.functionCallId
|
||||
: undefined
|
||||
}
|
||||
|
||||
const lowerToolCall = (part: ToolCallPart) => ({
|
||||
functionCall: { id: functionCallId(part.providerMetadata), name: part.name, args: part.input },
|
||||
thoughtSignature: thoughtSignature(part.providerMetadata),
|
||||
})
|
||||
|
||||
const lowerMessages = Effect.fn("Gemini.lowerMessages")(function* (request: LLMRequest) {
|
||||
const contents: GeminiContent[] = []
|
||||
|
||||
for (const message of request.messages) {
|
||||
if (message.role === "system") {
|
||||
const part = yield* ProviderShared.wrappedSystemUpdate("Gemini", message)
|
||||
const previous = contents.at(-1)
|
||||
if (previous?.role === "user")
|
||||
contents[contents.length - 1] = { role: "user", parts: [...previous.parts, { text: part.text }] }
|
||||
else contents.push({ role: "user", parts: [{ text: part.text }] })
|
||||
continue
|
||||
}
|
||||
|
||||
if (message.role === "user") {
|
||||
const parts: Array<Schema.Schema.Type<typeof GeminiContentPart>> = []
|
||||
for (const part of message.content) {
|
||||
if (!ProviderShared.supportsContent(part, ["text", "media"]))
|
||||
return yield* ProviderShared.unsupportedContent("Gemini", "user", ["text", "media"])
|
||||
parts.push(yield* lowerUserPart(part))
|
||||
}
|
||||
contents.push({ role: "user", parts })
|
||||
continue
|
||||
}
|
||||
|
||||
if (message.role === "assistant") {
|
||||
const parts: Array<Schema.Schema.Type<typeof GeminiContentPart>> = []
|
||||
// Parallel Gemini 3 calls may carry one signature on the first call; unsigned sibling calls are valid.
|
||||
let hasSignedToolCall = false
|
||||
for (const part of message.content) {
|
||||
if (!ProviderShared.supportsContent(part, ["text", "reasoning", "tool-call"]))
|
||||
return yield* ProviderShared.unsupportedContent("Gemini", "assistant", ["text", "reasoning", "tool-call"])
|
||||
if (part.type === "text") {
|
||||
parts.push({ text: part.text })
|
||||
continue
|
||||
}
|
||||
if (part.type === "reasoning") {
|
||||
parts.push({ text: part.text, thought: true, thoughtSignature: thoughtSignature(part.providerMetadata) })
|
||||
continue
|
||||
}
|
||||
if (part.type === "tool-call") {
|
||||
const lowered = lowerToolCall(part)
|
||||
const signature = lowered.thoughtSignature
|
||||
parts.push({
|
||||
...lowered,
|
||||
thoughtSignature:
|
||||
signature ??
|
||||
(requiresThoughtSignatureFallback(request.model.id) && !hasSignedToolCall
|
||||
? SKIP_THOUGHT_SIGNATURE_VALIDATOR
|
||||
: undefined),
|
||||
})
|
||||
if (signature !== undefined) hasSignedToolCall = true
|
||||
continue
|
||||
}
|
||||
}
|
||||
contents.push({ role: "model", parts })
|
||||
continue
|
||||
}
|
||||
|
||||
const parts: Array<Schema.Schema.Type<typeof GeminiContentPart>> = []
|
||||
for (const part of message.content) {
|
||||
if (!ProviderShared.supportsContent(part, ["tool-result"]))
|
||||
return yield* ProviderShared.unsupportedContent("Gemini", "tool", ["tool-result"])
|
||||
if (part.result.type !== "content") {
|
||||
parts.push({
|
||||
functionResponse: {
|
||||
id: functionCallId(part.providerMetadata),
|
||||
name: part.name,
|
||||
response: {
|
||||
name: part.name,
|
||||
content: ProviderShared.toolResultText(part),
|
||||
},
|
||||
},
|
||||
})
|
||||
continue
|
||||
}
|
||||
const content: ReadonlyArray<Tool.Content> = part.result.value
|
||||
const text = content.filter((item) => item.type === "text").map((item) => item.text)
|
||||
const media: GeminiInlineDataPart[] = []
|
||||
for (const item of content) {
|
||||
if (item.type === "text") continue
|
||||
const value = yield* ProviderShared.validateToolFile("Gemini", item, MEDIA_MIMES)
|
||||
media.push({ inlineData: { mimeType: value.mime, data: value.base64 } })
|
||||
}
|
||||
parts.push({
|
||||
functionResponse: {
|
||||
id: functionCallId(part.providerMetadata),
|
||||
name: part.name,
|
||||
response: {
|
||||
name: part.name,
|
||||
content: text.join("\n"),
|
||||
},
|
||||
parts: media.length > 0 ? media : undefined,
|
||||
},
|
||||
})
|
||||
}
|
||||
contents.push({ role: "user", parts })
|
||||
}
|
||||
|
||||
return contents
|
||||
})
|
||||
|
||||
const resolveOptions = (request: LLMRequest) => {
|
||||
const input = request.providerOptions?.gemini
|
||||
const value = input?.thinkingConfig
|
||||
const thinkingConfig = {
|
||||
thinkingBudget:
|
||||
ProviderShared.isRecord(value) && typeof value.thinkingBudget === "number" ? value.thinkingBudget : undefined,
|
||||
includeThoughts:
|
||||
ProviderShared.isRecord(value) && typeof value.includeThoughts === "boolean"
|
||||
? value.includeThoughts
|
||||
: ProviderShared.isRecord(value)
|
||||
? true
|
||||
: undefined,
|
||||
thinkingLevel:
|
||||
ProviderShared.isRecord(value) && typeof value.thinkingLevel === "string" ? value.thinkingLevel : undefined,
|
||||
}
|
||||
return {
|
||||
cachedContent: typeof input?.cachedContent === "string" ? input.cachedContent : undefined,
|
||||
safetySettings: mapSafetySettings(input?.safetySettings),
|
||||
serviceTier: typeof input?.serviceTier === "string" ? input.serviceTier : undefined,
|
||||
thinkingConfig: Object.values(thinkingConfig).some((item) => item !== undefined) ? thinkingConfig : undefined,
|
||||
}
|
||||
}
|
||||
|
||||
function mapSafetySettings(value: unknown) {
|
||||
if (!Array.isArray(value)) return undefined
|
||||
const settings = value.flatMap((item) =>
|
||||
ProviderShared.isRecord(item) && typeof item.category === "string" && typeof item.threshold === "string"
|
||||
? [{ category: item.category, threshold: item.threshold }]
|
||||
: [],
|
||||
)
|
||||
return settings
|
||||
}
|
||||
|
||||
const fromRequest = Effect.fn("Gemini.fromRequest")(function* (request: LLMRequest) {
|
||||
const hasTools = request.tools.length > 0
|
||||
const generation = request.generation
|
||||
const options = resolveOptions(request)
|
||||
const toolSchemaCompatibility = request.model.compatibility?.toolSchema
|
||||
const generationConfig = {
|
||||
maxOutputTokens: generation?.maxTokens,
|
||||
temperature: generation?.temperature,
|
||||
topP: generation?.topP,
|
||||
topK: generation?.topK,
|
||||
frequencyPenalty: generation?.frequencyPenalty,
|
||||
presencePenalty: generation?.presencePenalty,
|
||||
seed: generation?.seed,
|
||||
stopSequences: generation?.stop,
|
||||
thinkingConfig: options.thinkingConfig,
|
||||
}
|
||||
|
||||
return {
|
||||
cachedContent: options.cachedContent,
|
||||
contents: yield* lowerMessages(request),
|
||||
safetySettings: options.safetySettings,
|
||||
serviceTier: options.serviceTier,
|
||||
systemInstruction:
|
||||
request.system.length === 0 ? undefined : { parts: [{ text: ProviderShared.joinText(request.system) }] },
|
||||
tools: hasTools
|
||||
? [
|
||||
{
|
||||
functionDeclarations: request.tools.map((tool) =>
|
||||
lowerTool(tool, ToolSchemaProjection.modelCompatibility(tool.inputSchema, toolSchemaCompatibility)),
|
||||
),
|
||||
},
|
||||
]
|
||||
: undefined,
|
||||
toolConfig: hasTools && request.toolChoice ? yield* lowerToolConfig(request.toolChoice) : undefined,
|
||||
generationConfig: Object.values(generationConfig).some((value) => value !== undefined)
|
||||
? generationConfig
|
||||
: undefined,
|
||||
}
|
||||
})
|
||||
|
||||
// =============================================================================
|
||||
// Stream Parsing
|
||||
// =============================================================================
|
||||
// Gemini reports `promptTokenCount` (inclusive total) with a
|
||||
// `cachedContentTokenCount` subset. `candidatesTokenCount` is *exclusive*
|
||||
// of `thoughtsTokenCount` — visible-only, not a total — so we sum the two
|
||||
// to produce the inclusive `outputTokens` the rest of the contract expects.
|
||||
const mapUsage = (usage: GeminiUsage | undefined) => {
|
||||
if (!usage) return undefined
|
||||
const cached = usage.cachedContentTokenCount
|
||||
const nonCached = ProviderShared.subtractTokens(usage.promptTokenCount, cached)
|
||||
// `candidatesTokenCount` is visible-only; sum with thoughts to produce the
|
||||
// inclusive `outputTokens` the contract expects. Only compute the total
|
||||
// when the visible component is reported — otherwise we'd fabricate an
|
||||
// inclusive number from a partial breakdown.
|
||||
const outputTokens =
|
||||
usage.candidatesTokenCount !== undefined ? usage.candidatesTokenCount + (usage.thoughtsTokenCount ?? 0) : undefined
|
||||
return new Usage({
|
||||
inputTokens: usage.promptTokenCount,
|
||||
outputTokens,
|
||||
nonCachedInputTokens: nonCached,
|
||||
cacheReadInputTokens: cached,
|
||||
reasoningTokens: usage.thoughtsTokenCount,
|
||||
totalTokens: ProviderShared.totalTokens(usage.promptTokenCount, outputTokens, usage.totalTokenCount),
|
||||
providerMetadata: { google: usage },
|
||||
})
|
||||
}
|
||||
|
||||
const mapFinishReason = (finishReason: string | undefined, hasToolCalls: boolean): FinishReason => {
|
||||
if (finishReason === undefined) return hasToolCalls ? "tool-calls" : "unknown"
|
||||
if (finishReason === "STOP") return hasToolCalls ? "tool-calls" : "stop"
|
||||
if (finishReason === "MAX_TOKENS") return "length"
|
||||
if (
|
||||
finishReason === "IMAGE_SAFETY" ||
|
||||
finishReason === "RECITATION" ||
|
||||
finishReason === "SAFETY" ||
|
||||
finishReason === "BLOCKLIST" ||
|
||||
finishReason === "PROHIBITED_CONTENT" ||
|
||||
finishReason === "SPII" ||
|
||||
finishReason === "MODEL_ARMOR" ||
|
||||
finishReason === "IMAGE_PROHIBITED_CONTENT" ||
|
||||
finishReason === "IMAGE_RECITATION" ||
|
||||
finishReason === "LANGUAGE"
|
||||
)
|
||||
return "content-filter"
|
||||
if (
|
||||
finishReason === "MALFORMED_FUNCTION_CALL" ||
|
||||
finishReason === "UNEXPECTED_TOOL_CALL" ||
|
||||
finishReason === "NO_IMAGE" ||
|
||||
finishReason === "TOO_MANY_TOOL_CALLS" ||
|
||||
finishReason === "MISSING_THOUGHT_SIGNATURE" ||
|
||||
finishReason === "MALFORMED_RESPONSE"
|
||||
)
|
||||
return "error"
|
||||
return "unknown"
|
||||
}
|
||||
|
||||
const finish = (state: ParserState): ReadonlyArray<LLMEvent> =>
|
||||
state.finishReason || state.usage
|
||||
? (() => {
|
||||
const events: LLMEvent[] = []
|
||||
const lifecycle = state.reasoningSignature
|
||||
? Lifecycle.reasoningEnd(
|
||||
state.lifecycle,
|
||||
events,
|
||||
"reasoning-0",
|
||||
googleMetadata({ thoughtSignature: state.reasoningSignature }),
|
||||
)
|
||||
: state.lifecycle
|
||||
Lifecycle.finish(lifecycle, events, {
|
||||
reason: {
|
||||
normalized: mapFinishReason(state.finishReason, state.hasToolCalls),
|
||||
raw: state.finishReason,
|
||||
},
|
||||
usage: state.usage,
|
||||
})
|
||||
return events
|
||||
})()
|
||||
: []
|
||||
|
||||
const step = (state: ParserState, event: GeminiEvent) => {
|
||||
const nextState = {
|
||||
...state,
|
||||
usage: event.usageMetadata ? (mapUsage(event.usageMetadata) ?? state.usage) : state.usage,
|
||||
}
|
||||
const candidate = event.candidates?.[0]
|
||||
if (!candidate?.content)
|
||||
return Effect.succeed([
|
||||
{ ...nextState, finishReason: candidate?.finishReason ?? nextState.finishReason },
|
||||
[],
|
||||
] as const)
|
||||
|
||||
const events: LLMEvent[] = []
|
||||
let hasToolCalls = nextState.hasToolCalls
|
||||
let lifecycle = nextState.lifecycle
|
||||
let nextToolCallId = nextState.nextToolCallId
|
||||
let reasoningSignature = nextState.reasoningSignature
|
||||
|
||||
for (const part of candidate.content.parts) {
|
||||
if ("thoughtSignature" in part && part.thoughtSignature && "thought" in part && part.thought)
|
||||
reasoningSignature = part.thoughtSignature
|
||||
if ("text" in part && part.text.length > 0) {
|
||||
if (part.thought) {
|
||||
lifecycle = Lifecycle.reasoningDelta(
|
||||
lifecycle,
|
||||
events,
|
||||
"reasoning-0",
|
||||
part.text,
|
||||
part.thoughtSignature ? googleMetadata({ thoughtSignature: part.thoughtSignature }) : undefined,
|
||||
)
|
||||
continue
|
||||
}
|
||||
lifecycle = Lifecycle.reasoningEnd(
|
||||
lifecycle,
|
||||
events,
|
||||
"reasoning-0",
|
||||
reasoningSignature ? googleMetadata({ thoughtSignature: reasoningSignature }) : undefined,
|
||||
)
|
||||
lifecycle = Lifecycle.textDelta(lifecycle, events, "text-0", part.text)
|
||||
continue
|
||||
}
|
||||
|
||||
if ("functionCall" in part) {
|
||||
const input = part.functionCall.args
|
||||
const id = `tool_${nextToolCallId++}`
|
||||
const metadata = {
|
||||
...(part.functionCall.id === undefined ? {} : { functionCallId: part.functionCall.id }),
|
||||
...(part.thoughtSignature === undefined ? {} : { thoughtSignature: part.thoughtSignature }),
|
||||
}
|
||||
lifecycle = Lifecycle.reasoningEnd(
|
||||
lifecycle,
|
||||
events,
|
||||
"reasoning-0",
|
||||
reasoningSignature ? googleMetadata({ thoughtSignature: reasoningSignature }) : undefined,
|
||||
)
|
||||
lifecycle = Lifecycle.stepStart(lifecycle, events)
|
||||
events.push(
|
||||
LLMEvent.toolCall({
|
||||
id,
|
||||
name: part.functionCall.name,
|
||||
input,
|
||||
providerMetadata: Object.keys(metadata).length > 0 ? googleMetadata(metadata) : undefined,
|
||||
}),
|
||||
)
|
||||
hasToolCalls = true
|
||||
}
|
||||
}
|
||||
|
||||
return Effect.succeed([
|
||||
{
|
||||
...nextState,
|
||||
hasToolCalls,
|
||||
lifecycle,
|
||||
nextToolCallId,
|
||||
reasoningSignature,
|
||||
finishReason: candidate.finishReason ?? nextState.finishReason,
|
||||
},
|
||||
events,
|
||||
] as const)
|
||||
}
|
||||
|
||||
// =============================================================================
|
||||
// Protocol And Gemini Route
|
||||
// =============================================================================
|
||||
/**
|
||||
* The Gemini protocol — request body construction, body schema, and the
|
||||
* streaming-event state machine. Used by Google AI Studio Gemini and (once
|
||||
* registered) Vertex Gemini.
|
||||
*/
|
||||
export const protocol = Protocol.make({
|
||||
id: ADAPTER,
|
||||
body: {
|
||||
schema: GeminiBody,
|
||||
from: fromRequest,
|
||||
},
|
||||
stream: {
|
||||
event: Protocol.jsonEvent(GeminiEvent),
|
||||
initial: () => ({ hasToolCalls: false, nextToolCallId: 0, lifecycle: Lifecycle.initial() }),
|
||||
step,
|
||||
onHalt: finish,
|
||||
},
|
||||
})
|
||||
|
||||
export const route = Route.make({
|
||||
id: ADAPTER,
|
||||
provider: "google",
|
||||
providerMetadataKey: "google",
|
||||
protocol,
|
||||
// Gemini's path embeds the model id and pins SSE framing at the URL level.
|
||||
endpoint: Endpoint.path(({ request }) => `/models/${request.model.id}:streamGenerateContent?alt=sse`, {
|
||||
baseURL: DEFAULT_BASE_URL,
|
||||
}),
|
||||
auth: Auth.none,
|
||||
framing: Framing.sse,
|
||||
})
|
||||
|
||||
export * as Gemini from "./gemini.js"
|
||||
@@ -1,314 +0,0 @@
|
||||
import { Effect, Encoding, Schema } from "effect"
|
||||
import { Headers, HttpClientRequest } from "effect/unstable/http"
|
||||
import {
|
||||
GeneratedImage,
|
||||
ImageModel,
|
||||
ImageResponse,
|
||||
type ImageInput,
|
||||
type ImageRequestFor,
|
||||
type ImageRoute,
|
||||
} from "../image.js"
|
||||
import { Auth, type Definition as AuthDefinition } from "../route/auth.js"
|
||||
import {
|
||||
InvalidProviderOutputReason,
|
||||
AIError,
|
||||
Usage,
|
||||
mergeHttpOptions,
|
||||
mergeJsonRecords,
|
||||
type HttpOptions,
|
||||
type ProviderMetadata,
|
||||
} from "../schema/index.js"
|
||||
import { ProviderShared } from "./shared.js"
|
||||
import { ImageInputs } from "./utils/image-input.js"
|
||||
|
||||
const ADAPTER = "google-images"
|
||||
export const DEFAULT_BASE_URL = "https://generativelanguage.googleapis.com/v1beta"
|
||||
|
||||
export type GoogleImageString<Known extends string> = Known | (string & {})
|
||||
|
||||
export type GoogleImageOptions = {
|
||||
readonly aspectRatio?: GoogleImageString<
|
||||
"1:1" | "2:3" | "3:2" | "3:4" | "4:3" | "4:5" | "5:4" | "9:16" | "16:9" | "21:9"
|
||||
>
|
||||
readonly imageSize?: GoogleImageString<"1K" | "2K" | "4K">
|
||||
readonly seed?: number
|
||||
readonly thinkingLevel?: GoogleImageString<"MINIMAL" | "LOW" | "MEDIUM" | "HIGH">
|
||||
readonly includeThoughts?: boolean
|
||||
} & Record<string, unknown>
|
||||
|
||||
export type GoogleImageBody = Record<string, unknown> & {
|
||||
readonly contents: ReadonlyArray<{
|
||||
readonly role: "user"
|
||||
readonly parts: ReadonlyArray<Record<string, unknown>>
|
||||
}>
|
||||
readonly generationConfig: Record<string, unknown>
|
||||
}
|
||||
|
||||
const GoogleUsage = Schema.StructWithRest(
|
||||
Schema.Struct({
|
||||
cachedContentTokenCount: Schema.optional(Schema.Number),
|
||||
thoughtsTokenCount: Schema.optional(Schema.Number),
|
||||
promptTokenCount: Schema.optional(Schema.Number),
|
||||
candidatesTokenCount: Schema.optional(Schema.Number),
|
||||
totalTokenCount: Schema.optional(Schema.Number),
|
||||
promptTokensDetails: Schema.optional(Schema.Unknown),
|
||||
candidatesTokensDetails: Schema.optional(Schema.Unknown),
|
||||
}),
|
||||
[Schema.Record(Schema.String, Schema.Unknown)],
|
||||
)
|
||||
|
||||
const GoogleImageResponse = Schema.Struct({
|
||||
candidates: Schema.optional(
|
||||
Schema.Array(
|
||||
Schema.Struct({
|
||||
index: Schema.optional(Schema.Number),
|
||||
content: Schema.optional(
|
||||
Schema.Struct({
|
||||
parts: Schema.Array(
|
||||
Schema.Struct({
|
||||
text: Schema.optional(Schema.String),
|
||||
thought: Schema.optional(Schema.Boolean),
|
||||
thoughtSignature: Schema.optional(Schema.String),
|
||||
inlineData: Schema.optional(
|
||||
Schema.Struct({
|
||||
mimeType: Schema.String,
|
||||
data: Schema.String,
|
||||
}),
|
||||
),
|
||||
}),
|
||||
),
|
||||
}),
|
||||
),
|
||||
finishReason: Schema.optional(Schema.String),
|
||||
finishMessage: Schema.optional(Schema.String),
|
||||
safetyRatings: Schema.optional(Schema.Unknown),
|
||||
citationMetadata: Schema.optional(Schema.Unknown),
|
||||
groundingMetadata: Schema.optional(Schema.Unknown),
|
||||
}),
|
||||
),
|
||||
),
|
||||
usageMetadata: Schema.optional(GoogleUsage),
|
||||
modelVersion: Schema.optional(Schema.String),
|
||||
responseId: Schema.optional(Schema.String),
|
||||
promptFeedback: Schema.optional(Schema.Unknown),
|
||||
})
|
||||
|
||||
export interface ModelInput {
|
||||
readonly id: string
|
||||
readonly auth: AuthDefinition
|
||||
readonly baseURL?: string
|
||||
readonly headers?: Record<string, string>
|
||||
readonly http?: HttpOptions
|
||||
}
|
||||
|
||||
const nativeOptions = (options: GoogleImageOptions | undefined) => {
|
||||
const { aspectRatio, imageSize, seed, thinkingLevel, includeThoughts, ...native } = options ?? {}
|
||||
const image = {
|
||||
aspectRatio,
|
||||
imageSize,
|
||||
}
|
||||
const thinkingConfig = {
|
||||
thinkingLevel,
|
||||
includeThoughts,
|
||||
}
|
||||
return (
|
||||
mergeJsonRecords(
|
||||
{
|
||||
responseModalities: ["IMAGE"],
|
||||
imageConfig: Object.values(image).some((value) => value !== undefined) ? image : undefined,
|
||||
seed,
|
||||
thinkingConfig: Object.values(thinkingConfig).some((value) => value !== undefined) ? thinkingConfig : undefined,
|
||||
},
|
||||
native,
|
||||
) ?? { responseModalities: ["IMAGE"] }
|
||||
)
|
||||
}
|
||||
|
||||
const invalidOutput = (message: string, providerMetadata?: ProviderMetadata) =>
|
||||
new AIError({
|
||||
module: ADAPTER,
|
||||
method: "generate",
|
||||
reason: new InvalidProviderOutputReason({ message, route: ADAPTER, providerMetadata }),
|
||||
})
|
||||
|
||||
const applyQuery = (url: string, query: Record<string, string> | undefined) => {
|
||||
if (!query) return url
|
||||
const next = new URL(url)
|
||||
Object.entries(query).forEach(([key, value]) => next.searchParams.set(key, value))
|
||||
return next.toString()
|
||||
}
|
||||
|
||||
export const model = (input: ModelInput) => {
|
||||
const route: ImageRoute<GoogleImageOptions> = {
|
||||
id: ADAPTER,
|
||||
generate: Effect.fn("GoogleImages.generate")(function* (request: ImageRequestFor<GoogleImageOptions>, execute) {
|
||||
const imageParts = yield* Effect.forEach(request.images ?? [], googleImagePart)
|
||||
const http = mergeHttpOptions(request.model.http, request.http)
|
||||
const requestBody = mergeJsonRecords(
|
||||
{
|
||||
contents: [{ role: "user", parts: [{ text: request.prompt }, ...imageParts] }],
|
||||
generationConfig: nativeOptions(request.options),
|
||||
},
|
||||
http?.body,
|
||||
) as GoogleImageBody
|
||||
const text = ProviderShared.encodeJson(requestBody)
|
||||
const url = applyQuery(
|
||||
`${(input.baseURL ?? DEFAULT_BASE_URL).replace(/\/$/, "")}/models/${request.model.id}:generateContent`,
|
||||
http?.query,
|
||||
)
|
||||
const headers = yield* Auth.toEffect(input.auth)({
|
||||
request,
|
||||
method: "POST",
|
||||
url,
|
||||
body: text,
|
||||
headers: Headers.fromInput({ ...input.headers, ...http?.headers }),
|
||||
})
|
||||
const response = yield* execute(
|
||||
HttpClientRequest.post(url).pipe(
|
||||
HttpClientRequest.setHeaders(headers),
|
||||
HttpClientRequest.bodyText(text, "application/json"),
|
||||
),
|
||||
)
|
||||
const payload = yield* response.json.pipe(
|
||||
Effect.mapError(() => invalidOutput("Failed to read the Google Images response")),
|
||||
)
|
||||
const decoded = yield* Schema.decodeUnknownEffect(GoogleImageResponse)(payload).pipe(
|
||||
Effect.mapError(() => invalidOutput("Google Images returned an invalid response")),
|
||||
)
|
||||
const candidates = decoded.candidates ?? []
|
||||
const candidateMetadata = candidates.map((candidate, candidateIndex) => ({
|
||||
index: candidate.index ?? candidateIndex,
|
||||
finishReason: candidate.finishReason,
|
||||
finishMessage: candidate.finishMessage,
|
||||
safetyRatings: candidate.safetyRatings,
|
||||
citationMetadata: candidate.citationMetadata,
|
||||
groundingMetadata: candidate.groundingMetadata,
|
||||
parts: (candidate.content?.parts ?? []).map((part) =>
|
||||
part.inlineData === undefined
|
||||
? {
|
||||
type: "text",
|
||||
text: part.text,
|
||||
thought: part.thought,
|
||||
thoughtSignature: part.thoughtSignature,
|
||||
}
|
||||
: {
|
||||
type: "inlineData",
|
||||
mediaType: part.inlineData.mimeType,
|
||||
thought: part.thought,
|
||||
thoughtSignature: part.thoughtSignature,
|
||||
},
|
||||
),
|
||||
}))
|
||||
const encoded = candidates.flatMap((candidate, candidateIndex) =>
|
||||
(candidate.content?.parts ?? []).flatMap((part, partIndex) =>
|
||||
part.inlineData === undefined || part.thought === true
|
||||
? []
|
||||
: [{ candidate, candidateIndex, partIndex, inlineData: part.inlineData }],
|
||||
),
|
||||
)
|
||||
const images = yield* Effect.forEach(encoded, (item) =>
|
||||
Effect.fromResult(Encoding.decodeBase64(item.inlineData.data)).pipe(
|
||||
Effect.mapError(() =>
|
||||
invalidOutput(
|
||||
`Google Images candidate ${item.candidateIndex} part ${item.partIndex} contains invalid base64 data`,
|
||||
),
|
||||
),
|
||||
Effect.map(
|
||||
(data) =>
|
||||
new GeneratedImage({
|
||||
mediaType: item.inlineData.mimeType,
|
||||
data,
|
||||
providerMetadata: {
|
||||
google: {
|
||||
candidateIndex: item.candidate.index ?? item.candidateIndex,
|
||||
partIndex: item.partIndex,
|
||||
finishReason: item.candidate.finishReason,
|
||||
safetyRatings: item.candidate.safetyRatings,
|
||||
citationMetadata: item.candidate.citationMetadata,
|
||||
groundingMetadata: item.candidate.groundingMetadata,
|
||||
thoughtSignature: item.candidate.content?.parts[item.partIndex]?.thoughtSignature,
|
||||
},
|
||||
},
|
||||
}),
|
||||
),
|
||||
),
|
||||
)
|
||||
if (images.length === 0) {
|
||||
const finishReasons = candidates.flatMap((candidate) =>
|
||||
candidate.finishReason === undefined ? [] : [candidate.finishReason],
|
||||
)
|
||||
return yield* invalidOutput(
|
||||
`Google Images returned no final images${
|
||||
finishReasons.length === 0 ? "" : ` (finish reasons: ${finishReasons.join(", ")})`
|
||||
}; inspect reason.providerMetadata.google for prompt feedback and candidate details`,
|
||||
{
|
||||
google: {
|
||||
promptFeedback: decoded.promptFeedback,
|
||||
candidates: candidateMetadata,
|
||||
},
|
||||
},
|
||||
)
|
||||
}
|
||||
const usage = decoded.usageMetadata
|
||||
const outputTokens =
|
||||
usage?.candidatesTokenCount === undefined
|
||||
? undefined
|
||||
: usage.candidatesTokenCount + (usage.thoughtsTokenCount ?? 0)
|
||||
return new ImageResponse({
|
||||
images,
|
||||
usage:
|
||||
usage === undefined
|
||||
? undefined
|
||||
: new Usage({
|
||||
inputTokens: usage.promptTokenCount,
|
||||
outputTokens,
|
||||
nonCachedInputTokens: ProviderShared.subtractTokens(
|
||||
usage.promptTokenCount,
|
||||
usage.cachedContentTokenCount,
|
||||
),
|
||||
cacheReadInputTokens: usage.cachedContentTokenCount,
|
||||
reasoningTokens: usage.thoughtsTokenCount,
|
||||
totalTokens: ProviderShared.totalTokens(usage.promptTokenCount, outputTokens, usage.totalTokenCount),
|
||||
providerMetadata: { google: usage },
|
||||
}),
|
||||
providerMetadata: {
|
||||
google: {
|
||||
modelVersion: decoded.modelVersion,
|
||||
responseId: decoded.responseId,
|
||||
promptFeedback: decoded.promptFeedback,
|
||||
candidates: candidateMetadata,
|
||||
},
|
||||
},
|
||||
})
|
||||
}),
|
||||
}
|
||||
return ImageModel.make<GoogleImageOptions>({ id: input.id, provider: "google", route, http: input.http })
|
||||
}
|
||||
|
||||
const googleImagePart = (image: ImageInput): Effect.Effect<Record<string, unknown>, AIError> => {
|
||||
if (image.type === "bytes")
|
||||
return Effect.succeed({ inlineData: { mimeType: image.mediaType, data: Encoding.encodeBase64(image.data) } })
|
||||
if (image.type === "file-uri") return Effect.succeed({ fileData: { mimeType: image.mediaType, fileUri: image.uri } })
|
||||
if (image.type === "url")
|
||||
return ImageInputs.decodeDataUrl(image.url, ADAPTER).pipe(
|
||||
Effect.flatMap((decoded) => {
|
||||
if (decoded === undefined)
|
||||
return Effect.fail(
|
||||
ImageInputs.invalid(
|
||||
ADAPTER,
|
||||
"Google generateContent does not fetch public image URLs; use bytes, a data URL, or a Gemini file URI",
|
||||
),
|
||||
)
|
||||
return Effect.succeed({
|
||||
inlineData: { mimeType: decoded.mediaType, data: Encoding.encodeBase64(decoded.data) },
|
||||
})
|
||||
}),
|
||||
)
|
||||
return Effect.fail(
|
||||
ImageInputs.invalid(ADAPTER, "Google generateContent requires Gemini file URIs rather than provider file IDs"),
|
||||
)
|
||||
}
|
||||
|
||||
export const GoogleImages = {
|
||||
model,
|
||||
} as const
|
||||
@@ -1,10 +0,0 @@
|
||||
export * as AnthropicMessages from "./anthropic-messages.js"
|
||||
export * as BedrockConverse from "./bedrock-converse.js"
|
||||
export * as Gemini from "./gemini.js"
|
||||
export * as OpenAIChat from "./openai-chat.js"
|
||||
export * as OpenAIImages from "./openai-images.js"
|
||||
export * as OpenAICompatibleChat from "./openai-compatible-chat.js"
|
||||
export * as OpenAICompatibleResponses from "./openai-compatible-responses.js"
|
||||
export * as OpenAIResponses from "./openai-responses.js"
|
||||
export * as OpenResponses from "./open-responses.js"
|
||||
export * as OpenResponsesChannel from "./open-responses-channel.js"
|
||||
@@ -1,191 +0,0 @@
|
||||
import { Effect, Schema, Stream } from "effect"
|
||||
import { Headers } from "effect/unstable/http"
|
||||
import { Framing } from "../route/framing.js"
|
||||
import {
|
||||
HttpTransport,
|
||||
WebSocketTransport,
|
||||
type Transport,
|
||||
type WebSocketChannelDriver,
|
||||
type WebSocketChannelExchange,
|
||||
} from "../route/transport/index.js"
|
||||
import * as ProviderShared from "./shared.js"
|
||||
import { OpenResponses } from "./open-responses.js"
|
||||
|
||||
const WebSocketResponseCreate = Schema.StructWithRest(Schema.Struct({ type: Schema.tag("response.create") }), [
|
||||
Schema.Record(Schema.String, Schema.Unknown),
|
||||
])
|
||||
const decodeMessage = ProviderShared.validateWith(Schema.decodeUnknownEffect(WebSocketResponseCreate))
|
||||
const encodeMessage = Schema.encodeSync(Schema.fromJsonString(WebSocketResponseCreate))
|
||||
const decodeEvent = Schema.decodeUnknownEffect(OpenResponses.protocol.stream.event)
|
||||
|
||||
export interface Options {
|
||||
readonly id: string
|
||||
readonly name: string
|
||||
readonly rotateAfterMs?: number
|
||||
readonly headers?: (headers: Headers.Headers) => Headers.Headers
|
||||
readonly driver?: (input: {
|
||||
readonly request: Readonly<Record<string, unknown>>
|
||||
readonly message: string
|
||||
readonly base: WebSocketChannelDriver
|
||||
}) => WebSocketChannelDriver
|
||||
}
|
||||
|
||||
export interface Prepared {
|
||||
readonly http: HttpTransport.HttpPrepared<string>
|
||||
readonly channel?: {
|
||||
readonly url: string
|
||||
readonly headers: Headers.Headers
|
||||
readonly rotateAfterMs?: number
|
||||
readonly driver: WebSocketChannelDriver
|
||||
}
|
||||
}
|
||||
|
||||
const message = (body: unknown) =>
|
||||
Effect.gen(function* () {
|
||||
if (!ProviderShared.isRecord(body))
|
||||
return yield* ProviderShared.invalidRequest("Open Responses WebSocket body must be a JSON object")
|
||||
const { stream: _stream, stream_options: _streamOptions, background: _background, ...request } = body
|
||||
const decoded = yield* decodeMessage({ ...request, type: "response.create" })
|
||||
return { request: decoded, message: encodeMessage(decoded) }
|
||||
})
|
||||
|
||||
const driver = (options: Options, body: string): WebSocketChannelDriver => {
|
||||
let responseID: string | undefined
|
||||
let terminal = false
|
||||
return {
|
||||
create: () =>
|
||||
Effect.sync(() => {
|
||||
responseID = undefined
|
||||
terminal = false
|
||||
return { message: body, mode: "full" }
|
||||
}),
|
||||
observe: (_create, frame) =>
|
||||
Effect.gen(function* () {
|
||||
const event = yield* decodeEvent(frame).pipe(
|
||||
Effect.mapError(() =>
|
||||
ProviderShared.eventError(options.id, `Invalid ${options.name} WebSocket event`, frame),
|
||||
),
|
||||
)
|
||||
if (terminal)
|
||||
return yield* ProviderShared.eventError(
|
||||
options.id,
|
||||
`${options.name} emitted ${event.type} after a terminal event`,
|
||||
frame,
|
||||
)
|
||||
if (event.type === "error") {
|
||||
terminal = true
|
||||
yield* OpenResponses.decodeKnownErrorEvent(event).pipe(
|
||||
Effect.mapError(() =>
|
||||
ProviderShared.eventError(options.id, `${options.name} returned a malformed error event`, frame),
|
||||
),
|
||||
)
|
||||
return {
|
||||
type: "provider-failure",
|
||||
error: OpenResponses.providerFailure(options.id, event, `${options.name} stream error`),
|
||||
}
|
||||
}
|
||||
if (event.type === "response.failed") {
|
||||
terminal = true
|
||||
if (responseID && event.response?.id && event.response.id !== responseID)
|
||||
return yield* ProviderShared.eventError(
|
||||
options.id,
|
||||
`${options.name} response ID changed during execution`,
|
||||
frame,
|
||||
)
|
||||
return {
|
||||
type: "provider-failure",
|
||||
error: OpenResponses.providerFailure(options.id, event, `${options.name} response failed`),
|
||||
}
|
||||
}
|
||||
if (event.type === "response.created") {
|
||||
const created = event.response?.id
|
||||
if (responseID)
|
||||
return yield* ProviderShared.eventError(
|
||||
options.id,
|
||||
`${options.name} emitted duplicate response.created`,
|
||||
frame,
|
||||
)
|
||||
if (!created)
|
||||
return yield* ProviderShared.eventError(
|
||||
options.id,
|
||||
`${options.name} response.created is missing response.id`,
|
||||
frame,
|
||||
)
|
||||
responseID = created
|
||||
return { type: "frame", frame }
|
||||
}
|
||||
if (!responseID)
|
||||
return yield* ProviderShared.eventError(
|
||||
options.id,
|
||||
`${options.name} emitted ${event.type} before response.created`,
|
||||
frame,
|
||||
)
|
||||
if (event.response?.id && event.response.id !== responseID)
|
||||
return yield* ProviderShared.eventError(
|
||||
options.id,
|
||||
`${options.name} response ID changed during execution`,
|
||||
frame,
|
||||
)
|
||||
if (event.type === "response.completed") {
|
||||
terminal = true
|
||||
return { type: "completed", frame }
|
||||
}
|
||||
if (event.type === "response.incomplete") {
|
||||
terminal = true
|
||||
return { type: "incomplete", frame }
|
||||
}
|
||||
return { type: "frame", frame }
|
||||
}),
|
||||
}
|
||||
}
|
||||
|
||||
export const transport = <Body>(options: Options): Transport<Body, Prepared, string> => {
|
||||
const http = HttpTransport.sseJson.with<Body>()
|
||||
return {
|
||||
id: http.id,
|
||||
prepare: (input) =>
|
||||
Effect.gen(function* () {
|
||||
const parts = yield* HttpTransport.jsonRequestParts(input)
|
||||
const headers = Headers.remove(options.headers?.(parts.headers) ?? parts.headers, "content-length")
|
||||
const channel = input.webSocket
|
||||
? yield* Effect.gen(function* () {
|
||||
const create = yield* message(parts.jsonBody)
|
||||
const base = driver(options, create.message)
|
||||
return {
|
||||
url: yield* WebSocketTransport.toWebSocketUrl(parts.url),
|
||||
headers,
|
||||
rotateAfterMs: options.rotateAfterMs,
|
||||
driver: options.driver?.({ request: create.request, message: create.message, base }) ?? base,
|
||||
}
|
||||
})
|
||||
: undefined
|
||||
return {
|
||||
http: {
|
||||
request: ProviderShared.jsonPost({ url: parts.url, body: parts.bodyText, headers: parts.headers }),
|
||||
framing: Framing.sse,
|
||||
middleware: input.middleware,
|
||||
},
|
||||
channel,
|
||||
}
|
||||
}),
|
||||
execute: (prepared, request, runtime, executeOptions) => {
|
||||
if (!executeOptions?.webSocket || !prepared.channel) return http.execute(prepared.http, request, runtime)
|
||||
const exchange: WebSocketChannelExchange = {
|
||||
id: request.id ?? "request",
|
||||
connect: {
|
||||
url: prepared.channel.url,
|
||||
headers: prepared.channel.headers,
|
||||
rotateAfterMs: prepared.channel.rotateAfterMs,
|
||||
},
|
||||
fallback: () =>
|
||||
Stream.unwrap(
|
||||
http.execute(prepared.http, request, runtime).pipe(Effect.map((execution) => execution.frames)),
|
||||
),
|
||||
driver: prepared.channel.driver,
|
||||
}
|
||||
return executeOptions.webSocket.execute(exchange)
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
export const OpenResponsesChannel = { transport } as const
|
||||
File diff suppressed because it is too large
Load Diff
@@ -1,887 +0,0 @@
|
||||
import { Effect, Schema } from "effect"
|
||||
import { Tool } from "@opencode-ai/schema/tool"
|
||||
import { Route } from "../route/client.js"
|
||||
import { Auth } from "../route/auth.js"
|
||||
import { Endpoint } from "../route/endpoint.js"
|
||||
import { HttpTransport } from "../route/transport/index.js"
|
||||
import { Protocol } from "../route/protocol.js"
|
||||
import {
|
||||
AIError,
|
||||
LLMEvent,
|
||||
Usage,
|
||||
type FinishReason,
|
||||
type FinishReasonDetails,
|
||||
type CacheHint,
|
||||
type JsonSchema,
|
||||
type LLMRequest,
|
||||
type MediaPart,
|
||||
type ReasoningPart,
|
||||
type TextPart,
|
||||
type ToolCallPart,
|
||||
type ToolDefinition,
|
||||
} from "../schema/index.js"
|
||||
import { classifyProviderFailure } from "../provider-error.js"
|
||||
import { isRecord, JsonObject, optionalArray, optionalNull, ProviderShared } from "./shared.js"
|
||||
import { OpenAIOptions } from "./utils/openai-options.js"
|
||||
import { Lifecycle } from "./utils/lifecycle.js"
|
||||
import { ToolSchemaProjection } from "./utils/tool-schema.js"
|
||||
import { ToolStream } from "./utils/tool-stream.js"
|
||||
|
||||
const ADAPTER = "openai-chat"
|
||||
const IMAGE_MIMES = new Set<string>(ProviderShared.IMAGE_MIMES)
|
||||
const RESERVED_REASONING_FIELDS = new Set(["role", "content", "tool_calls"])
|
||||
export const DEFAULT_BASE_URL = "https://api.openai.com/v1"
|
||||
export const PATH = "/chat/completions"
|
||||
|
||||
// =============================================================================
|
||||
// Request Body Schema
|
||||
// =============================================================================
|
||||
// The body schema is the provider-native JSON body. `fromRequest` below builds
|
||||
// this shape from the common `LLMRequest`, then `Route.make` validates and
|
||||
// JSON-encodes it before transport.
|
||||
const OpenAIChatCacheControl = Schema.Struct({
|
||||
type: Schema.Literal("ephemeral"),
|
||||
ttl: Schema.optional(Schema.String),
|
||||
})
|
||||
|
||||
const OpenAIChatFunction = Schema.Struct({
|
||||
name: Schema.String,
|
||||
description: Schema.String,
|
||||
parameters: JsonObject,
|
||||
})
|
||||
|
||||
const OpenAIChatTool = Schema.Struct({
|
||||
type: Schema.tag("function"),
|
||||
function: OpenAIChatFunction,
|
||||
cache_control: Schema.optional(OpenAIChatCacheControl),
|
||||
})
|
||||
type OpenAIChatTool = Schema.Schema.Type<typeof OpenAIChatTool>
|
||||
|
||||
const OpenAIChatAssistantToolCall = Schema.Struct({
|
||||
id: Schema.String,
|
||||
type: Schema.tag("function"),
|
||||
function: Schema.Struct({
|
||||
name: Schema.String,
|
||||
arguments: Schema.String,
|
||||
}),
|
||||
})
|
||||
type OpenAIChatAssistantToolCall = Schema.Schema.Type<typeof OpenAIChatAssistantToolCall>
|
||||
|
||||
// Intentionally omit Gemini's provider-specific `extra_content.google.thought_signature`
|
||||
// extension until direct Google OpenAI-compatible routing is supported here:
|
||||
// https://github.com/vercel/ai/issues/11590
|
||||
// https://github.com/vercel/ai/pull/11745
|
||||
// https://ai.google.dev/gemini-api/docs/thought-signatures#openai
|
||||
|
||||
const OpenAIChatUserContent = Schema.Union([
|
||||
Schema.Struct({
|
||||
type: Schema.Literal("text"),
|
||||
text: Schema.String,
|
||||
cache_control: Schema.optional(OpenAIChatCacheControl),
|
||||
}),
|
||||
Schema.Struct({
|
||||
type: Schema.Literal("image_url"),
|
||||
image_url: Schema.Struct({ url: Schema.String }),
|
||||
}),
|
||||
])
|
||||
|
||||
const OpenAIChatMessage = Schema.Union([
|
||||
Schema.Struct({
|
||||
role: Schema.Literal("system"),
|
||||
content: Schema.Union([Schema.String, Schema.Array(OpenAIChatUserContent)]),
|
||||
}),
|
||||
Schema.Struct({
|
||||
role: Schema.Literal("user"),
|
||||
content: Schema.Union([Schema.String, Schema.Array(OpenAIChatUserContent)]),
|
||||
}),
|
||||
Schema.StructWithRest(
|
||||
Schema.Struct({
|
||||
role: Schema.Literal("assistant"),
|
||||
content: Schema.NullOr(Schema.String),
|
||||
tool_calls: optionalArray(OpenAIChatAssistantToolCall),
|
||||
reasoning_content: Schema.optional(Schema.String),
|
||||
reasoning: Schema.optional(Schema.String),
|
||||
reasoning_text: Schema.optional(Schema.String),
|
||||
reasoning_details: Schema.optional(Schema.Unknown),
|
||||
cache_control: Schema.optional(OpenAIChatCacheControl),
|
||||
}),
|
||||
[Schema.Record(Schema.String, Schema.Unknown)],
|
||||
),
|
||||
Schema.Struct({
|
||||
role: Schema.Literal("tool"),
|
||||
tool_call_id: Schema.String,
|
||||
content: Schema.String,
|
||||
cache_control: Schema.optional(OpenAIChatCacheControl),
|
||||
}),
|
||||
]).pipe(Schema.toTaggedUnion("role"))
|
||||
type OpenAIChatMessage = Schema.Schema.Type<typeof OpenAIChatMessage>
|
||||
|
||||
const OpenAIChatToolChoice = Schema.Union([
|
||||
Schema.Literals(["auto", "none", "required"]),
|
||||
Schema.Struct({
|
||||
type: Schema.tag("function"),
|
||||
function: Schema.Struct({ name: Schema.String }),
|
||||
}),
|
||||
])
|
||||
|
||||
export const bodyFields = {
|
||||
model: Schema.String,
|
||||
messages: Schema.Array(OpenAIChatMessage),
|
||||
tools: optionalArray(OpenAIChatTool),
|
||||
tool_choice: Schema.optional(OpenAIChatToolChoice),
|
||||
stream: Schema.Literal(true),
|
||||
stream_options: Schema.optional(Schema.Struct({ include_usage: Schema.Boolean })),
|
||||
store: Schema.optional(Schema.Boolean),
|
||||
prompt_cache_key: Schema.optional(Schema.String),
|
||||
reasoning_effort: Schema.optional(OpenAIOptions.OpenAIReasoningEffort),
|
||||
max_completion_tokens: Schema.optional(Schema.Number),
|
||||
max_tokens: Schema.optional(Schema.Number),
|
||||
temperature: Schema.optional(Schema.Number),
|
||||
top_p: Schema.optional(Schema.Number),
|
||||
frequency_penalty: Schema.optional(Schema.Number),
|
||||
presence_penalty: Schema.optional(Schema.Number),
|
||||
seed: Schema.optional(Schema.Number),
|
||||
stop: optionalArray(Schema.String),
|
||||
}
|
||||
const OpenAIChatBody = Schema.Struct(bodyFields)
|
||||
export type OpenAIChatBody = Schema.Schema.Type<typeof OpenAIChatBody>
|
||||
|
||||
// =============================================================================
|
||||
// Streaming Event Schema
|
||||
// =============================================================================
|
||||
// The event schema is one decoded SSE `data:` payload. `Framing.sse` splits the
|
||||
// byte stream into strings, then `Protocol.jsonEvent` decodes each string into
|
||||
// this provider-native event shape.
|
||||
const OpenAIChatUsage = Schema.StructWithRest(
|
||||
Schema.Struct({
|
||||
prompt_tokens: optionalNull(Schema.Number),
|
||||
completion_tokens: optionalNull(Schema.Number),
|
||||
total_tokens: optionalNull(Schema.Number),
|
||||
prompt_tokens_details: optionalNull(
|
||||
Schema.StructWithRest(
|
||||
Schema.Struct({
|
||||
cached_tokens: optionalNull(Schema.Number),
|
||||
cache_write_tokens: optionalNull(Schema.Number),
|
||||
}),
|
||||
[Schema.Record(Schema.String, Schema.Unknown)],
|
||||
),
|
||||
),
|
||||
completion_tokens_details: optionalNull(
|
||||
Schema.StructWithRest(
|
||||
Schema.Struct({
|
||||
reasoning_tokens: optionalNull(Schema.Number),
|
||||
accepted_prediction_tokens: optionalNull(Schema.Number),
|
||||
rejected_prediction_tokens: optionalNull(Schema.Number),
|
||||
}),
|
||||
[Schema.Record(Schema.String, Schema.Unknown)],
|
||||
),
|
||||
),
|
||||
}),
|
||||
[Schema.Record(Schema.String, Schema.Unknown)],
|
||||
)
|
||||
|
||||
const OpenAIChatToolCallDeltaFunction = Schema.Struct({
|
||||
name: optionalNull(Schema.String),
|
||||
arguments: optionalNull(Schema.String),
|
||||
})
|
||||
|
||||
const OpenAIChatToolCallDelta = Schema.Struct({
|
||||
index: optionalNull(Schema.Number),
|
||||
id: optionalNull(Schema.String),
|
||||
function: optionalNull(OpenAIChatToolCallDeltaFunction),
|
||||
})
|
||||
type OpenAIChatToolCallDelta = Schema.Schema.Type<typeof OpenAIChatToolCallDelta>
|
||||
|
||||
const OpenAIChatDelta = Schema.StructWithRest(
|
||||
Schema.Struct({
|
||||
content: optionalNull(Schema.String),
|
||||
reasoning_content: optionalNull(Schema.String),
|
||||
reasoning: optionalNull(Schema.String),
|
||||
reasoning_text: optionalNull(Schema.String),
|
||||
reasoning_details: optionalNull(Schema.Unknown),
|
||||
tool_calls: optionalNull(Schema.Array(OpenAIChatToolCallDelta)),
|
||||
}),
|
||||
[Schema.Record(Schema.String, Schema.Unknown)],
|
||||
)
|
||||
|
||||
const OpenAIChatChoice = Schema.Struct({
|
||||
delta: optionalNull(OpenAIChatDelta),
|
||||
finish_reason: optionalNull(Schema.String),
|
||||
native_finish_reason: optionalNull(Schema.String),
|
||||
})
|
||||
|
||||
const OpenAIChatError = Schema.Struct({
|
||||
code: optionalNull(Schema.Union([Schema.String, Schema.Number])),
|
||||
message: Schema.String,
|
||||
})
|
||||
|
||||
export const OpenAIChatEvent = Schema.Struct({
|
||||
choices: optionalNull(Schema.Array(OpenAIChatChoice)),
|
||||
usage: optionalNull(OpenAIChatUsage),
|
||||
error: optionalNull(OpenAIChatError),
|
||||
})
|
||||
export type OpenAIChatEvent = Schema.Schema.Type<typeof OpenAIChatEvent>
|
||||
type OpenAIChatRequestMessage = LLMRequest["messages"][number]
|
||||
|
||||
interface PendingToolDelta {
|
||||
readonly id?: string
|
||||
readonly name?: string
|
||||
readonly input: string
|
||||
}
|
||||
|
||||
export interface ParserState {
|
||||
readonly tools: ToolStream.State<number>
|
||||
readonly pendingTools: Partial<Record<number, PendingToolDelta>>
|
||||
readonly toolCallEvents: ReadonlyArray<LLMEvent>
|
||||
readonly usage?: Usage
|
||||
readonly finishReason?: FinishReasonDetails
|
||||
readonly lifecycle: Lifecycle.State
|
||||
readonly reasoningField?: string
|
||||
readonly reasoningDetails: Array<unknown>
|
||||
readonly reasoningDetailsObserved: boolean
|
||||
readonly reasoningEmitted: boolean
|
||||
readonly latestToolIndex?: number
|
||||
readonly nextToolIndex: number
|
||||
}
|
||||
|
||||
// =============================================================================
|
||||
// Request Lowering
|
||||
// =============================================================================
|
||||
// Lowering is the only place that knows how common LLM messages map onto the
|
||||
// OpenAI Chat wire format. Keep provider quirks here instead of leaking native
|
||||
// fields into `LLMRequest`.
|
||||
interface LoweringOptions {
|
||||
readonly cacheControl?: (
|
||||
cache: CacheHint | undefined,
|
||||
) => Schema.Schema.Type<typeof OpenAIChatCacheControl> | undefined
|
||||
}
|
||||
|
||||
const lowerTool = (tool: ToolDefinition, inputSchema: JsonSchema, options: LoweringOptions): OpenAIChatTool => ({
|
||||
type: "function",
|
||||
function: {
|
||||
name: tool.name,
|
||||
description: tool.description,
|
||||
parameters: ToolSchemaProjection.openAI(inputSchema),
|
||||
},
|
||||
cache_control: options.cacheControl?.(tool.cache),
|
||||
})
|
||||
|
||||
const lowerToolChoice = (toolChoice: NonNullable<LLMRequest["toolChoice"]>) =>
|
||||
ProviderShared.matchToolChoice("OpenAI Chat", toolChoice, {
|
||||
auto: () => "auto" as const,
|
||||
none: () => "none" as const,
|
||||
required: () => "required" as const,
|
||||
tool: (name) => ({ type: "function" as const, function: { name } }),
|
||||
})
|
||||
|
||||
const lowerToolCall = (part: ToolCallPart): OpenAIChatAssistantToolCall => ({
|
||||
id: part.id,
|
||||
type: "function",
|
||||
function: {
|
||||
name: part.name,
|
||||
arguments: ProviderShared.encodeJson(part.input),
|
||||
},
|
||||
})
|
||||
|
||||
const lowerMedia = Effect.fn("OpenAIChat.lowerMedia")(function* (part: MediaPart) {
|
||||
const media = yield* ProviderShared.validateMedia("OpenAI Chat", part, IMAGE_MIMES)
|
||||
return { type: "image_url" as const, image_url: { url: media.dataUrl } }
|
||||
})
|
||||
|
||||
const openAICompatibleReasoningContent = (native: unknown) =>
|
||||
isRecord(native) && typeof native.reasoning_content === "string" ? native.reasoning_content : undefined
|
||||
|
||||
const reasoningField = (part: ReasoningPart) => {
|
||||
const field = part.providerMetadata?.openai?.reasoningField
|
||||
return typeof field === "string" ? field : undefined
|
||||
}
|
||||
|
||||
const reasoningDetails = (parts: ReadonlyArray<ReasoningPart>, native: unknown) => {
|
||||
const observed = parts.flatMap((part) => {
|
||||
const details = part.providerMetadata?.openai?.reasoningDetails
|
||||
return Array.isArray(details) ? details : []
|
||||
})
|
||||
if (parts.some((part) => Array.isArray(part.providerMetadata?.openai?.reasoningDetails))) return observed
|
||||
if (isRecord(native) && Array.isArray(native.reasoning_details)) return native.reasoning_details
|
||||
}
|
||||
|
||||
const lowerUserMessage = Effect.fn("OpenAIChat.lowerUserMessage")(function* (
|
||||
message: OpenAIChatRequestMessage,
|
||||
options: LoweringOptions,
|
||||
) {
|
||||
const content: Array<Schema.Schema.Type<typeof OpenAIChatUserContent>> = []
|
||||
for (const part of message.content) {
|
||||
if (part.type === "text") {
|
||||
content.push({ type: "text", text: part.text, cache_control: options.cacheControl?.(part.cache) })
|
||||
continue
|
||||
}
|
||||
if (part.type === "media") {
|
||||
content.push(yield* lowerMedia(part))
|
||||
continue
|
||||
}
|
||||
return yield* ProviderShared.unsupportedContent("OpenAI Chat", "user", ["text", "media"])
|
||||
}
|
||||
if (content.every((part) => part.type === "text" && part.cache_control === undefined))
|
||||
return {
|
||||
role: "user" as const,
|
||||
content: content.map((part) => (part.type === "text" ? part.text : "")).join(""),
|
||||
}
|
||||
return { role: "user" as const, content }
|
||||
})
|
||||
|
||||
const lowerAssistantMessage = Effect.fn("OpenAIChat.lowerAssistantMessage")(function* (
|
||||
message: OpenAIChatRequestMessage,
|
||||
configuredField?: string,
|
||||
options: LoweringOptions = {},
|
||||
) {
|
||||
const content: TextPart[] = []
|
||||
const reasoning: ReasoningPart[] = []
|
||||
const toolCalls: OpenAIChatAssistantToolCall[] = []
|
||||
for (const part of message.content) {
|
||||
if (!ProviderShared.supportsContent(part, ["text", "reasoning", "tool-call"]))
|
||||
return yield* ProviderShared.unsupportedContent("OpenAI Chat", "assistant", ["text", "reasoning", "tool-call"])
|
||||
if (part.type === "text") {
|
||||
content.push(part)
|
||||
continue
|
||||
}
|
||||
if (part.type === "reasoning") {
|
||||
reasoning.push(part)
|
||||
continue
|
||||
}
|
||||
if (part.type === "tool-call") {
|
||||
toolCalls.push(lowerToolCall(part))
|
||||
continue
|
||||
}
|
||||
}
|
||||
const text = reasoning.map((part) => part.text).join("")
|
||||
const details = reasoningDetails(reasoning, message.native?.openaiCompatible)
|
||||
const observedField = reasoning.map(reasoningField).find((value) => value !== undefined)
|
||||
const nativeReasoning = openAICompatibleReasoningContent(message.native?.openaiCompatible)
|
||||
const fullyStructured = reasoning.every((part) => Array.isArray(part.providerMetadata?.openai?.reasoningDetails))
|
||||
const field = (() => {
|
||||
if (configuredField !== undefined) return configuredField
|
||||
if (reasoning.length === 0) return undefined
|
||||
if (observedField !== undefined) return observedField
|
||||
if (nativeReasoning !== undefined) return "reasoning_content"
|
||||
if (!fullyStructured) return "reasoning_content"
|
||||
})()
|
||||
const reasoningText = (() => {
|
||||
if (configuredField !== undefined) return reasoning.length === 0 ? (nativeReasoning ?? "") : text
|
||||
if (reasoning.length === 0) return nativeReasoning
|
||||
return text
|
||||
})()
|
||||
const cached = message.content.findLast((part) => "cache" in part && part.cache !== undefined)
|
||||
const cacheControl = options.cacheControl?.(cached && "cache" in cached ? cached.cache : undefined)
|
||||
const result = {
|
||||
role: "assistant" as const,
|
||||
content: content.length > 0 ? content.map((part) => part.text).join("") : toolCalls.length > 0 ? null : "",
|
||||
...(toolCalls.length > 0 ? { tool_calls: toolCalls } : {}),
|
||||
...(details !== undefined ? { reasoning_details: details } : {}),
|
||||
...(cacheControl !== undefined ? { cache_control: cacheControl } : {}),
|
||||
}
|
||||
if (field === undefined || reasoningText === undefined) return result
|
||||
return { ...result, [field]: reasoningText }
|
||||
})
|
||||
|
||||
const lowerToolMessages = Effect.fn("OpenAIChat.lowerToolMessages")(function* (
|
||||
message: OpenAIChatRequestMessage,
|
||||
options: LoweringOptions,
|
||||
) {
|
||||
const messages: OpenAIChatMessage[] = []
|
||||
const images: Array<Schema.Schema.Type<typeof OpenAIChatUserContent>> = []
|
||||
for (const part of message.content) {
|
||||
if (!ProviderShared.supportsContent(part, ["tool-result"]))
|
||||
return yield* ProviderShared.unsupportedContent("OpenAI Chat", "tool", ["tool-result"])
|
||||
if (part.result.type !== "content") {
|
||||
messages.push({
|
||||
role: "tool",
|
||||
tool_call_id: part.id,
|
||||
content: ProviderShared.toolResultText(part),
|
||||
cache_control: options.cacheControl?.(part.cache),
|
||||
})
|
||||
continue
|
||||
}
|
||||
const content: ReadonlyArray<Tool.Content> = part.result.value
|
||||
const text = content.filter((item) => item.type === "text").map((item) => item.text)
|
||||
messages.push({
|
||||
role: "tool",
|
||||
tool_call_id: part.id,
|
||||
content: text.join("\n"),
|
||||
cache_control: options.cacheControl?.(part.cache),
|
||||
})
|
||||
const files = content.filter((item) => item.type === "file")
|
||||
images.push(
|
||||
...(yield* Effect.forEach(files, (item) =>
|
||||
lowerMedia({ type: "media", mediaType: item.mime, data: item.uri, filename: item.name }),
|
||||
)),
|
||||
)
|
||||
}
|
||||
return { messages, images }
|
||||
})
|
||||
|
||||
const lowerMessage = Effect.fn("OpenAIChat.lowerMessage")(function* (
|
||||
message: OpenAIChatRequestMessage,
|
||||
reasoningField?: string,
|
||||
options: LoweringOptions = {},
|
||||
) {
|
||||
if (message.role === "user") return [yield* lowerUserMessage(message, options)]
|
||||
if (message.role === "assistant") return [yield* lowerAssistantMessage(message, reasoningField, options)]
|
||||
return (yield* lowerToolMessages(message, options)).messages
|
||||
})
|
||||
|
||||
const lowerMessages = Effect.fn("OpenAIChat.lowerMessages")(function* (request: LLMRequest, options: LoweringOptions) {
|
||||
const system: OpenAIChatMessage[] =
|
||||
request.system.length === 0
|
||||
? []
|
||||
: request.system.some((part) => part.cache !== undefined) && options.cacheControl !== undefined
|
||||
? [
|
||||
{
|
||||
role: "system",
|
||||
content: request.system.map((part) => ({
|
||||
type: "text",
|
||||
text: part.text,
|
||||
cache_control: options.cacheControl?.(part.cache),
|
||||
})),
|
||||
},
|
||||
]
|
||||
: [{ role: "system", content: ProviderShared.joinText(request.system) }]
|
||||
const messages = [...system]
|
||||
const pendingImages: Array<Schema.Schema.Type<typeof OpenAIChatUserContent>> = []
|
||||
const flushImages = () => {
|
||||
if (pendingImages.length === 0) return
|
||||
messages.push({ role: "user", content: pendingImages.splice(0) })
|
||||
}
|
||||
for (const message of request.messages) {
|
||||
if (message.role === "system") {
|
||||
const part = yield* ProviderShared.wrappedSystemUpdate("OpenAI Chat", message)
|
||||
if (pendingImages.length > 0) {
|
||||
messages.push({
|
||||
role: "user",
|
||||
content: [
|
||||
...pendingImages.splice(0),
|
||||
{ type: "text", text: part.text, cache_control: options.cacheControl?.(part.cache) },
|
||||
],
|
||||
})
|
||||
continue
|
||||
}
|
||||
const previous = messages.at(-1)
|
||||
if (previous?.role === "user" && typeof previous.content === "string")
|
||||
messages[messages.length - 1] = options.cacheControl?.(part.cache)
|
||||
? {
|
||||
role: "user",
|
||||
content: [
|
||||
{ type: "text", text: previous.content },
|
||||
{ type: "text", text: part.text, cache_control: options.cacheControl(part.cache) },
|
||||
],
|
||||
}
|
||||
: { role: "user", content: `${previous.content}\n${part.text}` }
|
||||
else if (previous?.role === "user" && Array.isArray(previous.content))
|
||||
messages[messages.length - 1] = {
|
||||
role: "user",
|
||||
content: [
|
||||
...previous.content,
|
||||
{ type: "text", text: part.text, cache_control: options.cacheControl?.(part.cache) },
|
||||
],
|
||||
}
|
||||
else
|
||||
messages.push(
|
||||
options.cacheControl?.(part.cache)
|
||||
? {
|
||||
role: "user",
|
||||
content: [{ type: "text", text: part.text, cache_control: options.cacheControl(part.cache) }],
|
||||
}
|
||||
: { role: "user", content: part.text },
|
||||
)
|
||||
continue
|
||||
}
|
||||
if (message.role === "tool") {
|
||||
const lowered = yield* lowerToolMessages(message, options)
|
||||
messages.push(...lowered.messages)
|
||||
pendingImages.push(...lowered.images)
|
||||
continue
|
||||
}
|
||||
flushImages()
|
||||
messages.push(...(yield* lowerMessage(message, request.model.compatibility?.reasoningField, options)))
|
||||
}
|
||||
flushImages()
|
||||
return messages
|
||||
})
|
||||
|
||||
const lowerOptions = (request: LLMRequest) => {
|
||||
const options = OpenAIOptions.resolve(request)
|
||||
return {
|
||||
...(options.store !== undefined ? { store: options.store } : {}),
|
||||
...(request.promptCacheKey ? { prompt_cache_key: request.promptCacheKey } : {}),
|
||||
...(options.reasoningEffort ? { reasoning_effort: options.reasoningEffort } : {}),
|
||||
}
|
||||
}
|
||||
|
||||
export const fromRequest = Effect.fn("OpenAIChat.fromRequest")(function* (
|
||||
request: LLMRequest,
|
||||
options: LoweringOptions = {},
|
||||
) {
|
||||
// `fromRequest` returns the provider body only. Endpoint, auth, framing,
|
||||
// validation, and HTTP execution are composed by `Route.make`.
|
||||
const reasoningField = request.model.compatibility?.reasoningField
|
||||
if (reasoningField && RESERVED_REASONING_FIELDS.has(reasoningField))
|
||||
return yield* ProviderShared.invalidRequest(
|
||||
`OpenAI Chat reasoning field conflicts with reserved field ${reasoningField}`,
|
||||
)
|
||||
const generation = request.generation
|
||||
const toolSchemaCompatibility = request.model.compatibility?.toolSchema
|
||||
const maxTokensField = request.model.compatibility?.maxTokensField ?? "max_tokens"
|
||||
return {
|
||||
model: request.model.id,
|
||||
messages: yield* lowerMessages(request, options),
|
||||
tools:
|
||||
request.tools.length === 0
|
||||
? undefined
|
||||
: request.tools.map((tool) =>
|
||||
lowerTool(
|
||||
tool,
|
||||
ToolSchemaProjection.modelCompatibility(tool.inputSchema, toolSchemaCompatibility),
|
||||
options,
|
||||
),
|
||||
),
|
||||
tool_choice: request.toolChoice ? yield* lowerToolChoice(request.toolChoice) : undefined,
|
||||
stream: true as const,
|
||||
stream_options: { include_usage: true },
|
||||
...(maxTokensField === "max_completion_tokens"
|
||||
? { max_completion_tokens: generation?.maxTokens }
|
||||
: { max_tokens: generation?.maxTokens }),
|
||||
temperature: generation?.temperature,
|
||||
top_p: generation?.topP,
|
||||
frequency_penalty: generation?.frequencyPenalty,
|
||||
presence_penalty: generation?.presencePenalty,
|
||||
seed: generation?.seed,
|
||||
stop: generation?.stop,
|
||||
...lowerOptions(request),
|
||||
}
|
||||
})
|
||||
|
||||
// =============================================================================
|
||||
// Stream Parsing
|
||||
// =============================================================================
|
||||
// Streaming parsers are small state machines: every event returns a new state
|
||||
// plus the common `LLMEvent`s produced by that event. Tool calls are accumulated
|
||||
// because OpenAI streams JSON arguments across multiple deltas.
|
||||
const mapFinishReason = (reason: string | null | undefined): FinishReason => {
|
||||
if (reason === "stop") return "stop"
|
||||
if (reason === "length") return "length"
|
||||
if (reason === "content_filter") return "content-filter"
|
||||
if (reason === "function_call" || reason === "tool_calls") return "tool-calls"
|
||||
if (reason === "error") return "error"
|
||||
return "unknown"
|
||||
}
|
||||
|
||||
// OpenAI Chat reports `prompt_tokens` (inclusive total) with a
|
||||
// cached-read and cache-write subsets, and `completion_tokens` (inclusive
|
||||
// total) with a `reasoning_tokens` subset. We pass the inclusive totals
|
||||
// through and derive the non-cached breakdown so the `AI.Usage` contract is
|
||||
// satisfied on both sides.
|
||||
const mapUsage = (usage: OpenAIChatEvent["usage"]): Usage | undefined => {
|
||||
if (!usage) return undefined
|
||||
const input = usage.prompt_tokens ?? undefined
|
||||
const output = usage.completion_tokens ?? undefined
|
||||
const cached = usage.prompt_tokens_details?.cached_tokens ?? undefined
|
||||
const cacheWrite = usage.prompt_tokens_details?.cache_write_tokens ?? undefined
|
||||
const reasoning = usage.completion_tokens_details?.reasoning_tokens ?? undefined
|
||||
const nonCached = ProviderShared.subtractTokens(input, ProviderShared.sumTokens(cached, cacheWrite))
|
||||
return new Usage({
|
||||
inputTokens: input,
|
||||
outputTokens: output,
|
||||
nonCachedInputTokens: nonCached,
|
||||
cacheReadInputTokens: cached,
|
||||
cacheWriteInputTokens: cacheWrite,
|
||||
reasoningTokens: reasoning,
|
||||
totalTokens: ProviderShared.totalTokens(input, output, usage.total_tokens ?? undefined),
|
||||
providerMetadata: { openai: usage },
|
||||
})
|
||||
}
|
||||
|
||||
const toolIndexByID = (
|
||||
tools: ParserState["tools"],
|
||||
pendingTools: ParserState["pendingTools"],
|
||||
id: string | undefined,
|
||||
) => {
|
||||
if (!id) return undefined
|
||||
const entry = Object.entries({ ...pendingTools, ...tools }).find(([, tool]) => tool?.id === id)
|
||||
return entry ? Number(entry[0]) : undefined
|
||||
}
|
||||
|
||||
const reasoningDelta = (
|
||||
delta: Schema.Schema.Type<typeof OpenAIChatDelta> | null | undefined,
|
||||
configuredField?: string,
|
||||
) => {
|
||||
if (!delta) return undefined
|
||||
const fields = new Set([configuredField, "reasoning_content", "reasoning", "reasoning_text"])
|
||||
for (const field of fields) {
|
||||
if (field === undefined) continue
|
||||
const text = delta[field]
|
||||
if (typeof text === "string" && text.length > 0) return { field, text }
|
||||
}
|
||||
return undefined
|
||||
}
|
||||
|
||||
const detailText = (details: ReadonlyArray<unknown>) => {
|
||||
const text = details.flatMap((detail) => {
|
||||
if (!isRecord(detail)) return []
|
||||
if (detail.type === "reasoning.text" && typeof detail.text === "string" && detail.text) return [detail.text]
|
||||
if (detail.type === "reasoning.summary" && typeof detail.summary === "string" && detail.summary)
|
||||
return [detail.summary]
|
||||
return []
|
||||
})
|
||||
if (text.length > 0) return text.join("")
|
||||
}
|
||||
|
||||
const appendReasoningDetails = (result: Array<unknown>, details: ReadonlyArray<unknown>) => {
|
||||
for (const detail of details) {
|
||||
const previous = result.at(-1)
|
||||
if (
|
||||
!isRecord(previous) ||
|
||||
previous.type !== "reasoning.text" ||
|
||||
!isRecord(detail) ||
|
||||
detail.type !== "reasoning.text" ||
|
||||
conflictingReasoningTextDetails(previous, detail)
|
||||
) {
|
||||
result.push(detail)
|
||||
continue
|
||||
}
|
||||
result[result.length - 1] = {
|
||||
...previous,
|
||||
...Object.fromEntries(Object.entries(detail).filter((entry) => entry[1] !== undefined)),
|
||||
text: `${typeof previous.text === "string" ? previous.text : ""}${typeof detail.text === "string" ? detail.text : ""}`,
|
||||
signature: mergeDetailValue(previous.signature, detail.signature),
|
||||
format: mergeDetailValue(previous.format, detail.format),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
const mergeDetailValue = (previous: unknown, current: unknown) =>
|
||||
previous || current || (previous !== undefined ? previous : current)
|
||||
|
||||
const conflictingReasoningTextDetails = (previous: Record<string, unknown>, current: Record<string, unknown>) =>
|
||||
conflictingDetailValue(previous.id, current.id) ||
|
||||
conflictingDetailValue(previous.index, current.index) ||
|
||||
conflictingDetailValue(previous.format, current.format) ||
|
||||
(Boolean(previous.signature) && Boolean(current.signature) && previous.signature !== current.signature)
|
||||
|
||||
const conflictingDetailValue = (previous: unknown, current: unknown) =>
|
||||
previous !== undefined && previous !== null && current !== undefined && current !== null && previous !== current
|
||||
|
||||
const reasoningMetadata = (field: ParserState["reasoningField"], details?: ReadonlyArray<unknown>) => ({
|
||||
openai: {
|
||||
...(field ? { reasoningField: field } : {}),
|
||||
...(details ? { reasoningDetails: details } : {}),
|
||||
},
|
||||
})
|
||||
|
||||
const step = (state: ParserState, event: OpenAIChatEvent) =>
|
||||
Effect.gen(function* () {
|
||||
if (event.error)
|
||||
return yield* new AIError({
|
||||
module: ADAPTER,
|
||||
method: "stream",
|
||||
reason: classifyProviderFailure({
|
||||
message: event.error.message,
|
||||
code: event.error.code === undefined || event.error.code === null ? undefined : String(event.error.code),
|
||||
status: typeof event.error.code === "number" ? event.error.code : undefined,
|
||||
}),
|
||||
})
|
||||
const events: LLMEvent[] = []
|
||||
const usage = mapUsage(event.usage) ?? state.usage
|
||||
const choice = event.choices?.[0]
|
||||
const rawFinishReason = choice?.finish_reason
|
||||
const finishReason =
|
||||
rawFinishReason !== undefined && rawFinishReason !== null
|
||||
? { normalized: mapFinishReason(rawFinishReason), raw: choice?.native_finish_reason ?? rawFinishReason }
|
||||
: state.finishReason
|
||||
const delta = choice?.delta
|
||||
const toolDeltas = delta?.tool_calls ?? []
|
||||
let tools = state.tools
|
||||
let pendingTools = state.pendingTools
|
||||
let latestToolIndex = state.latestToolIndex
|
||||
let nextToolIndex = state.nextToolIndex
|
||||
|
||||
let lifecycle = state.lifecycle
|
||||
|
||||
const reasoning = reasoningDelta(delta, state.reasoningField)
|
||||
const hasLateContent =
|
||||
Boolean(delta?.content) ||
|
||||
reasoning !== undefined ||
|
||||
(Array.isArray(delta?.reasoning_details) && delta.reasoning_details.length > 0) ||
|
||||
toolDeltas.some((tool) => Boolean(tool.id) || Boolean(tool.function?.name) || Boolean(tool.function?.arguments))
|
||||
if (state.finishReason !== undefined) {
|
||||
if (hasLateContent)
|
||||
return yield* ProviderShared.eventError(ADAPTER, "OpenAI Chat received content after the finish reason")
|
||||
return [{ ...state, usage }, events] as const
|
||||
}
|
||||
|
||||
const reasoningField = state.reasoningField ?? reasoning?.field
|
||||
const detailDelta = Array.isArray(delta?.reasoning_details) ? delta.reasoning_details : undefined
|
||||
if (detailDelta !== undefined) appendReasoningDetails(state.reasoningDetails, detailDelta)
|
||||
const reasoningDetailsObserved = state.reasoningDetailsObserved || detailDelta !== undefined
|
||||
const deltaMetadata = reasoningMetadata(reasoningField)
|
||||
const text = detailDelta?.length ? (detailText(detailDelta) ?? reasoning?.text) : reasoning?.text
|
||||
if (text !== undefined) lifecycle = Lifecycle.reasoningDelta(lifecycle, events, "reasoning-0", text, deltaMetadata)
|
||||
else if (
|
||||
reasoningDetailsObserved &&
|
||||
!lifecycle.reasoning.has("reasoning-0") &&
|
||||
(Boolean(delta?.content) || toolDeltas.length > 0)
|
||||
)
|
||||
lifecycle = Lifecycle.reasoningStart(lifecycle, events, "reasoning-0", deltaMetadata)
|
||||
const reasoningEmitted = state.reasoningEmitted || lifecycle.reasoning.has("reasoning-0")
|
||||
|
||||
if (delta?.content) {
|
||||
lifecycle = Lifecycle.reasoningEnd(
|
||||
lifecycle,
|
||||
events,
|
||||
"reasoning-0",
|
||||
reasoningMetadata(reasoningField, reasoningDetailsObserved ? state.reasoningDetails : undefined),
|
||||
)
|
||||
lifecycle = Lifecycle.textDelta(lifecycle, events, "text-0", delta.content)
|
||||
}
|
||||
|
||||
// Compatible providers may omit indexes. Prefer durable identity, then use
|
||||
// batch position for parallel deltas or the latest call for sparse chunks.
|
||||
for (const [position, tool] of toolDeltas.entries()) {
|
||||
const matched = toolIndexByID(tools, pendingTools, tool.id || undefined)
|
||||
const fallback = toolDeltas.length > 1 ? position : (latestToolIndex ?? position)
|
||||
const fallbackTool = tools[fallback] ?? pendingTools[fallback]
|
||||
const index =
|
||||
tool.index ?? matched ?? (tool.id && fallbackTool?.id && fallbackTool.id !== tool.id ? nextToolIndex : fallback)
|
||||
const current = tools[index]
|
||||
const pending = pendingTools[index]
|
||||
const id = current?.id ?? pending?.id ?? (tool.id || undefined)
|
||||
const name = current?.name ?? pending?.name ?? (tool.function?.name || undefined)
|
||||
const text = `${pending?.input ?? ""}${tool.function?.arguments ?? ""}`
|
||||
latestToolIndex = index
|
||||
nextToolIndex = Math.max(nextToolIndex, index + 1)
|
||||
if (!current && (!id || !name)) {
|
||||
pendingTools = {
|
||||
...pendingTools,
|
||||
[index]: { id: id || undefined, name: name || undefined, input: text },
|
||||
}
|
||||
continue
|
||||
}
|
||||
if (pending) {
|
||||
pendingTools = { ...pendingTools }
|
||||
delete pendingTools[index]
|
||||
}
|
||||
const result = ToolStream.appendOrStart(
|
||||
ADAPTER,
|
||||
tools,
|
||||
index,
|
||||
{ id: id || undefined, name: name || undefined, text },
|
||||
"OpenAI Chat tool call delta is missing id or name",
|
||||
)
|
||||
if (ToolStream.isError(result)) return yield* result
|
||||
tools = result.tools
|
||||
if (result.events.length) lifecycle = Lifecycle.stepStart(lifecycle, events)
|
||||
events.push(...result.events)
|
||||
}
|
||||
|
||||
if (finishReason !== undefined && state.finishReason === undefined && Object.keys(pendingTools).length > 0)
|
||||
return yield* ProviderShared.eventError(ADAPTER, "OpenAI Chat tool call delta is missing id or name")
|
||||
|
||||
// Finalize accumulated tool inputs eagerly when finish_reason arrives so
|
||||
// valid calls and malformed local calls settle independently.
|
||||
const finished =
|
||||
finishReason !== undefined && state.finishReason === undefined && Object.keys(tools).length > 0
|
||||
? yield* ToolStream.finishAll(ADAPTER, tools)
|
||||
: undefined
|
||||
|
||||
return [
|
||||
{
|
||||
tools: finished?.tools ?? tools,
|
||||
pendingTools,
|
||||
toolCallEvents: finished?.events ?? state.toolCallEvents,
|
||||
usage,
|
||||
finishReason,
|
||||
lifecycle,
|
||||
reasoningField,
|
||||
reasoningDetails: state.reasoningDetails,
|
||||
reasoningDetailsObserved,
|
||||
reasoningEmitted,
|
||||
latestToolIndex,
|
||||
nextToolIndex,
|
||||
},
|
||||
events,
|
||||
] as const
|
||||
})
|
||||
|
||||
const finishEvents = (state: ParserState): ReadonlyArray<LLMEvent> => {
|
||||
const events: LLMEvent[] = []
|
||||
const toolCallEvents =
|
||||
state.finishReason === undefined && Object.keys(state.tools).length > 0
|
||||
? Effect.runSync(ToolStream.finishAll(ADAPTER, state.tools)).events
|
||||
: state.toolCallEvents
|
||||
const hasToolCalls = toolCallEvents.length > 0
|
||||
const reason = state.finishReason
|
||||
? {
|
||||
...state.finishReason,
|
||||
normalized:
|
||||
state.finishReason.normalized === "stop" && hasToolCalls ? "tool-calls" : state.finishReason.normalized,
|
||||
}
|
||||
: { normalized: hasToolCalls ? ("tool-calls" as const) : ("unknown" as const) }
|
||||
const metadata = reasoningMetadata(
|
||||
state.reasoningField,
|
||||
state.reasoningDetailsObserved ? state.reasoningDetails : undefined,
|
||||
)
|
||||
const started =
|
||||
state.reasoningDetailsObserved && !state.reasoningEmitted
|
||||
? Lifecycle.reasoningStart(state.lifecycle, events, "reasoning-0", reasoningMetadata(state.reasoningField))
|
||||
: state.lifecycle
|
||||
const ended = Lifecycle.reasoningEnd(started, events, "reasoning-0", metadata)
|
||||
const lifecycle = toolCallEvents.length ? Lifecycle.stepStart(ended, events) : ended
|
||||
events.push(...toolCallEvents)
|
||||
Lifecycle.finish(lifecycle, events, { reason, usage: state.usage })
|
||||
return events
|
||||
}
|
||||
|
||||
// =============================================================================
|
||||
// Protocol And OpenAI Route
|
||||
// =============================================================================
|
||||
/**
|
||||
* The OpenAI Chat protocol — request body construction, body schema, and the
|
||||
* streaming-event state machine. Reused by every route that speaks OpenAI Chat
|
||||
* over HTTP+SSE: native OpenAI, DeepSeek, TogetherAI, Cerebras, Baseten,
|
||||
* Fireworks, DeepInfra, and (once added) Azure OpenAI Chat.
|
||||
*/
|
||||
export const protocol = Protocol.make({
|
||||
id: ADAPTER,
|
||||
body: {
|
||||
schema: OpenAIChatBody,
|
||||
from: fromRequest,
|
||||
},
|
||||
stream: {
|
||||
event: Protocol.jsonEvent(OpenAIChatEvent),
|
||||
initial: (request) => ({
|
||||
tools: ToolStream.empty<number>(),
|
||||
pendingTools: {},
|
||||
toolCallEvents: [],
|
||||
lifecycle: Lifecycle.initial(),
|
||||
reasoningField: request.model.compatibility?.reasoningField,
|
||||
reasoningDetails: [],
|
||||
reasoningDetailsObserved: false,
|
||||
reasoningEmitted: false,
|
||||
nextToolIndex: 0,
|
||||
}),
|
||||
step,
|
||||
onHalt: finishEvents,
|
||||
},
|
||||
})
|
||||
|
||||
export const httpTransport = HttpTransport.sseJson.with<OpenAIChatBody>()
|
||||
|
||||
export const route = Route.make({
|
||||
id: ADAPTER,
|
||||
provider: "openai",
|
||||
providerMetadataKey: "openai",
|
||||
protocol,
|
||||
endpoint: Endpoint.path(PATH, { baseURL: DEFAULT_BASE_URL }),
|
||||
auth: Auth.none,
|
||||
transport: httpTransport,
|
||||
})
|
||||
|
||||
export * as OpenAIChat from "./openai-chat.js"
|
||||
@@ -1,25 +0,0 @@
|
||||
import { Route, type RouteRoutedLanguageModelInput } from "../route/client.js"
|
||||
import { Endpoint } from "../route/endpoint.js"
|
||||
import { Framing } from "../route/framing.js"
|
||||
import * as OpenAIChat from "./openai-chat.js"
|
||||
|
||||
const ADAPTER = "openai-compatible-chat"
|
||||
|
||||
export type OpenAICompatibleChatLanguageModelInput = RouteRoutedLanguageModelInput
|
||||
|
||||
/**
|
||||
* Route for non-OpenAI providers that expose an OpenAI Chat-compatible
|
||||
* `/chat/completions` endpoint. Reuses `OpenAIChat.protocol` end-to-end and
|
||||
* overrides only the route id so providers can be resolved per-family without
|
||||
* colliding with native OpenAI. Provider helpers configure the route endpoint
|
||||
* before model selection.
|
||||
*/
|
||||
export const route = Route.make({
|
||||
id: ADAPTER,
|
||||
providerMetadataKey: "openai",
|
||||
protocol: OpenAIChat.protocol,
|
||||
endpoint: Endpoint.path("/chat/completions"),
|
||||
framing: Framing.sse,
|
||||
})
|
||||
|
||||
export * as OpenAICompatibleChat from "./openai-compatible-chat.js"
|
||||
@@ -1,22 +0,0 @@
|
||||
import { Route, type RouteRoutedLanguageModelInput } from "../route/client.js"
|
||||
import { Endpoint } from "../route/endpoint.js"
|
||||
import { OpenResponses } from "./open-responses.js"
|
||||
|
||||
const ADAPTER = "openai-compatible-responses"
|
||||
|
||||
export type OpenAICompatibleResponsesLanguageModelInput = RouteRoutedLanguageModelInput
|
||||
|
||||
/**
|
||||
* Deployment adapter for providers that expose an Open Responses-compatible
|
||||
* `/responses` endpoint. Provider helpers configure identity, endpoint, and
|
||||
* auth while the semantic protocol remains provider-neutral.
|
||||
*/
|
||||
export const route = Route.make({
|
||||
id: ADAPTER,
|
||||
providerMetadataKey: "openresponses",
|
||||
protocol: OpenResponses.protocol,
|
||||
endpoint: Endpoint.path(OpenResponses.PATH),
|
||||
transport: OpenResponses.httpTransport,
|
||||
})
|
||||
|
||||
export * as OpenAICompatibleResponses from "./openai-compatible-responses.js"
|
||||
@@ -1,270 +0,0 @@
|
||||
import { Effect, Encoding, Schema } from "effect"
|
||||
import { Headers, HttpClientRequest, HttpClientResponse } from "effect/unstable/http"
|
||||
import {
|
||||
ImageModel,
|
||||
GeneratedImage,
|
||||
ImageResponse,
|
||||
type ImageInput,
|
||||
type ImageRequestFor,
|
||||
type ImageRoute,
|
||||
} from "../image.js"
|
||||
import { Auth, type Definition as AuthDefinition } from "../route/auth.js"
|
||||
import {
|
||||
InvalidProviderOutputReason,
|
||||
AIError,
|
||||
Usage,
|
||||
mergeHttpOptions,
|
||||
mergeJsonRecords,
|
||||
type HttpOptions,
|
||||
} from "../schema/index.js"
|
||||
import { ProviderShared } from "./shared.js"
|
||||
import { ImageInputs } from "./utils/image-input.js"
|
||||
import { OpenAIImage } from "./utils/openai-image.js"
|
||||
|
||||
const ADAPTER = "openai-images"
|
||||
export const DEFAULT_BASE_URL = "https://api.openai.com/v1"
|
||||
export const PATH = "/images/generations"
|
||||
export const EDIT_PATH = "/images/edits"
|
||||
|
||||
export type OpenAIImageString<Known extends string> = Known | (string & {})
|
||||
|
||||
export type OpenAIImageOptions = {
|
||||
readonly mask?: ImageInput
|
||||
readonly n?: number
|
||||
readonly size?: OpenAIImageString<
|
||||
"auto" | "256x256" | "512x512" | "1024x1024" | "1536x1024" | "1024x1536" | "1792x1024" | "1024x1792"
|
||||
>
|
||||
readonly quality?: OpenAIImageString<"auto" | "low" | "medium" | "high" | "standard" | "hd">
|
||||
readonly background?: OpenAIImageString<"auto" | "opaque" | "transparent">
|
||||
readonly moderation?: OpenAIImageString<"auto" | "low">
|
||||
readonly outputFormat?: OpenAIImageString<"png" | "jpeg" | "webp">
|
||||
readonly outputCompression?: number
|
||||
} & Record<string, unknown>
|
||||
|
||||
export type OpenAIImageBody = Record<string, unknown> & {
|
||||
readonly model: string
|
||||
readonly prompt: string
|
||||
}
|
||||
|
||||
const OpenAIImageResponse = Schema.Struct({
|
||||
data: Schema.Array(
|
||||
Schema.Struct({
|
||||
b64_json: Schema.optional(Schema.String),
|
||||
url: Schema.optional(Schema.String),
|
||||
revised_prompt: Schema.optional(Schema.String),
|
||||
}),
|
||||
),
|
||||
output_format: Schema.optional(Schema.String),
|
||||
usage: Schema.optional(
|
||||
Schema.Struct({
|
||||
input_tokens: Schema.optional(Schema.Number),
|
||||
output_tokens: Schema.optional(Schema.Number),
|
||||
total_tokens: Schema.optional(Schema.Number),
|
||||
input_tokens_details: Schema.optional(Schema.Record(Schema.String, Schema.Unknown)),
|
||||
output_tokens_details: Schema.optional(Schema.Record(Schema.String, Schema.Unknown)),
|
||||
}),
|
||||
),
|
||||
})
|
||||
|
||||
export interface ModelInput {
|
||||
readonly id: string
|
||||
readonly auth: AuthDefinition
|
||||
readonly baseURL?: string
|
||||
readonly headers?: Record<string, string>
|
||||
readonly http?: HttpOptions
|
||||
}
|
||||
|
||||
const nativeOptions = (options: OpenAIImageOptions | undefined) => {
|
||||
if (!options) return undefined
|
||||
const { mask: _, outputFormat, outputCompression, ...native } = options
|
||||
return {
|
||||
output_format: outputFormat,
|
||||
output_compression: outputCompression,
|
||||
...native,
|
||||
}
|
||||
}
|
||||
|
||||
const invalidOutput = (message: string) =>
|
||||
new AIError({
|
||||
module: ADAPTER,
|
||||
method: "generate",
|
||||
reason: new InvalidProviderOutputReason({ message, route: ADAPTER }),
|
||||
})
|
||||
|
||||
const applyQuery = (url: string, query: Record<string, string> | undefined) => {
|
||||
if (!query) return url
|
||||
const next = new URL(url)
|
||||
Object.entries(query).forEach(([key, value]) => next.searchParams.set(key, value))
|
||||
return next.toString()
|
||||
}
|
||||
|
||||
export const model = (input: ModelInput) => {
|
||||
const route: ImageRoute<OpenAIImageOptions> = {
|
||||
id: ADAPTER,
|
||||
generate: Effect.fn("OpenAIImages.generate")(function* (request: ImageRequestFor<OpenAIImageOptions>, execute) {
|
||||
const mask = request.options?.mask
|
||||
if (mask !== undefined && (request.images?.length ?? 0) === 0)
|
||||
return yield* ImageInputs.invalid(ADAPTER, "An OpenAI image mask requires at least one input image")
|
||||
const http = mergeHttpOptions(request.model.http, request.http)
|
||||
const sourceImages = request.images ?? []
|
||||
const multipartImages = yield* Effect.forEach(sourceImages, (image) => {
|
||||
if (image.type === "bytes") return Effect.succeed({ data: image.data, mediaType: image.mediaType })
|
||||
if (image.type === "url") return ImageInputs.decodeDataUrl(image.url, ADAPTER)
|
||||
return Effect.succeed(undefined)
|
||||
})
|
||||
const multipartMask =
|
||||
mask === undefined
|
||||
? undefined
|
||||
: mask.type === "bytes"
|
||||
? { data: mask.data, mediaType: mask.mediaType }
|
||||
: mask.type === "url"
|
||||
? yield* ImageInputs.decodeDataUrl(mask.url, ADAPTER)
|
||||
: undefined
|
||||
const useMultipart =
|
||||
sourceImages.length > 0 &&
|
||||
multipartImages.every((image) => image !== undefined) &&
|
||||
(mask === undefined || multipartMask !== undefined)
|
||||
const path = sourceImages.length === 0 ? PATH : EDIT_PATH
|
||||
const url = applyQuery(`${(input.baseURL ?? DEFAULT_BASE_URL).replace(/\/$/, "")}${path}`, http?.query)
|
||||
|
||||
if (useMultipart) {
|
||||
const form = new FormData()
|
||||
form.append("model", request.model.id)
|
||||
form.append("prompt", request.prompt)
|
||||
Object.entries(mergeJsonRecords(nativeOptions(request.options), http?.body) ?? {}).forEach(([key, value]) => {
|
||||
if (["model", "prompt", "image", "image[]", "images", "mask"].includes(key)) return
|
||||
form.append(key, typeof value === "string" ? value : ProviderShared.encodeJson(value))
|
||||
})
|
||||
multipartImages.forEach((image, index) => {
|
||||
if (image === undefined) return
|
||||
form.append("image[]", imageBlob(image.data, image.mediaType), `image-${index}`)
|
||||
})
|
||||
if (multipartMask !== undefined)
|
||||
form.append("mask", imageBlob(multipartMask.data, multipartMask.mediaType), "mask")
|
||||
const headers = yield* Auth.toEffect(input.auth)({
|
||||
request,
|
||||
method: "POST",
|
||||
url,
|
||||
body: "[multipart/form-data]",
|
||||
headers: Headers.remove(Headers.fromInput({ ...input.headers, ...http?.headers }), "content-type"),
|
||||
})
|
||||
const response = yield* execute(
|
||||
HttpClientRequest.post(url).pipe(HttpClientRequest.setHeaders(headers), HttpClientRequest.bodyFormData(form)),
|
||||
)
|
||||
return yield* parseResponse(response, request.options, http?.body)
|
||||
}
|
||||
|
||||
const references = sourceImages.map((image) => {
|
||||
if (image.type === "bytes") return { image_url: ImageInputs.dataUrl(image) }
|
||||
if (image.type === "url") return { image_url: image.url }
|
||||
if (image.type === "file-id") return { file_id: image.id }
|
||||
return undefined
|
||||
})
|
||||
if (references.some((image) => image === undefined))
|
||||
return yield* ImageInputs.invalid(ADAPTER, "OpenAI Images accepts image URLs, data URLs, bytes, and file IDs")
|
||||
const maskReference =
|
||||
mask === undefined
|
||||
? undefined
|
||||
: mask.type === "bytes"
|
||||
? { image_url: ImageInputs.dataUrl(mask) }
|
||||
: mask.type === "url"
|
||||
? { image_url: mask.url }
|
||||
: mask.type === "file-id"
|
||||
? { file_id: mask.id }
|
||||
: undefined
|
||||
if (mask !== undefined && maskReference === undefined)
|
||||
return yield* ImageInputs.invalid(ADAPTER, "OpenAI Images accepts masks as URLs, data URLs, bytes, or file IDs")
|
||||
const requestBody = mergeJsonRecords(
|
||||
{
|
||||
model: request.model.id,
|
||||
prompt: request.prompt,
|
||||
images: references.length === 0 ? undefined : references,
|
||||
mask: maskReference,
|
||||
},
|
||||
nativeOptions(request.options),
|
||||
http?.body,
|
||||
) as OpenAIImageBody
|
||||
const text = ProviderShared.encodeJson(requestBody)
|
||||
const headers = yield* Auth.toEffect(input.auth)({
|
||||
request,
|
||||
method: "POST",
|
||||
url,
|
||||
body: text,
|
||||
headers: Headers.fromInput({ ...input.headers, ...http?.headers }),
|
||||
})
|
||||
const response = yield* execute(
|
||||
HttpClientRequest.post(url).pipe(
|
||||
HttpClientRequest.setHeaders(headers),
|
||||
HttpClientRequest.bodyText(text, "application/json"),
|
||||
),
|
||||
)
|
||||
return yield* parseResponse(response, request.options, http?.body)
|
||||
}),
|
||||
}
|
||||
return ImageModel.make<OpenAIImageOptions>({ id: input.id, provider: "openai", route, http: input.http })
|
||||
}
|
||||
|
||||
const parseResponse = Effect.fn("OpenAIImages.parseResponse")(function* (
|
||||
response: HttpClientResponse.HttpClientResponse,
|
||||
options: OpenAIImageOptions | undefined,
|
||||
overlay: Record<string, unknown> | undefined,
|
||||
) {
|
||||
const payload = yield* response.json.pipe(
|
||||
Effect.mapError(() => invalidOutput("Failed to read the OpenAI Images response")),
|
||||
)
|
||||
const decoded = yield* Schema.decodeUnknownEffect(OpenAIImageResponse)(payload).pipe(
|
||||
Effect.mapError(() => invalidOutput("OpenAI Images returned an invalid response")),
|
||||
)
|
||||
const requestBody = mergeJsonRecords(nativeOptions(options), overlay)
|
||||
const format =
|
||||
decoded.output_format ?? (typeof requestBody?.output_format === "string" ? requestBody.output_format : "png")
|
||||
const images = yield* Effect.forEach(decoded.data, (item, index) => {
|
||||
if (item.b64_json)
|
||||
return Effect.fromResult(Encoding.decodeBase64(item.b64_json)).pipe(
|
||||
Effect.mapError(() => invalidOutput(`OpenAI Images result ${index} contains invalid base64 data`)),
|
||||
Effect.map(
|
||||
(data) =>
|
||||
new GeneratedImage({
|
||||
mediaType: `image/${format}`,
|
||||
data,
|
||||
providerMetadata:
|
||||
item.revised_prompt === undefined ? undefined : { openai: { revisedPrompt: item.revised_prompt } },
|
||||
}),
|
||||
),
|
||||
)
|
||||
if (item.url)
|
||||
return Effect.succeed(
|
||||
new GeneratedImage({
|
||||
mediaType: `image/${format}`,
|
||||
data: item.url,
|
||||
providerMetadata:
|
||||
item.revised_prompt === undefined ? undefined : { openai: { revisedPrompt: item.revised_prompt } },
|
||||
}),
|
||||
)
|
||||
return Effect.fail(invalidOutput(`OpenAI Images result ${index} has neither image data nor a URL`))
|
||||
})
|
||||
if (images.length === 0) return yield* invalidOutput("OpenAI Images returned no images")
|
||||
return new ImageResponse({
|
||||
images,
|
||||
usage:
|
||||
decoded.usage === undefined
|
||||
? undefined
|
||||
: new Usage({
|
||||
inputTokens: decoded.usage.input_tokens,
|
||||
outputTokens: decoded.usage.output_tokens,
|
||||
totalTokens: decoded.usage.total_tokens,
|
||||
providerMetadata: { openai: decoded.usage },
|
||||
}),
|
||||
providerMetadata: { openai: { outputFormat: format } },
|
||||
})
|
||||
})
|
||||
|
||||
const imageBlob = (data: Uint8Array, mediaType: string) => {
|
||||
const buffer = new ArrayBuffer(data.byteLength)
|
||||
new Uint8Array(buffer).set(data)
|
||||
return new Blob([buffer], { type: mediaType })
|
||||
}
|
||||
|
||||
export const OpenAIImages = {
|
||||
model,
|
||||
} as const
|
||||
@@ -1,164 +0,0 @@
|
||||
import { AIError, TransportReason } from "../schema/index.js"
|
||||
import type { ChannelCheckpoint, ChannelObservation, WebSocketChannelDriver } from "../route/transport/index.js"
|
||||
import { Effect, Option, Schema } from "effect"
|
||||
import * as ProviderShared from "./shared.js"
|
||||
import { OpenResponses } from "./open-responses.js"
|
||||
|
||||
const PROTOCOL = "openai-responses.websocket.v1"
|
||||
const VERSION = 1
|
||||
const decodeEvent = Schema.decodeUnknownEffect(OpenResponses.protocol.stream.event)
|
||||
|
||||
interface CheckpointValue {
|
||||
readonly version: typeof VERSION
|
||||
readonly responseID: string
|
||||
readonly request: Readonly<Record<string, unknown>>
|
||||
readonly output: ReadonlyArray<unknown>
|
||||
}
|
||||
|
||||
export interface DriverInput {
|
||||
readonly id: string
|
||||
readonly name: string
|
||||
readonly request: Readonly<Record<string, unknown>>
|
||||
readonly message: string
|
||||
readonly base: WebSocketChannelDriver
|
||||
}
|
||||
|
||||
const checkpointValue = (checkpoint: ChannelCheckpoint | undefined): CheckpointValue | undefined => {
|
||||
if (checkpoint?.protocol !== PROTOCOL || !ProviderShared.isRecord(checkpoint.value)) return undefined
|
||||
if (checkpoint.value.version !== VERSION) return undefined
|
||||
if (typeof checkpoint.value.responseID !== "string" || checkpoint.value.responseID.trim().length === 0)
|
||||
return undefined
|
||||
if (!ProviderShared.isRecord(checkpoint.value.request) || !Array.isArray(checkpoint.value.output)) return undefined
|
||||
return {
|
||||
version: VERSION,
|
||||
responseID: checkpoint.value.responseID,
|
||||
request: checkpoint.value.request,
|
||||
output: checkpoint.value.output,
|
||||
}
|
||||
}
|
||||
|
||||
const canonical = (value: unknown): string => {
|
||||
if (value === undefined) return "undefined"
|
||||
if (Array.isArray(value)) return `[${value.map(canonical).join(",")}]`
|
||||
if (!ProviderShared.isRecord(value)) return ProviderShared.encodeJson(value)
|
||||
return `{${Object.keys(value)
|
||||
.sort()
|
||||
.map((key) => `${ProviderShared.encodeJson(key)}:${canonical(value[key])}`)
|
||||
.join(",")}}`
|
||||
}
|
||||
|
||||
const json = (value: unknown) => {
|
||||
if (typeof value !== "string") return value
|
||||
return Option.getOrElse(Schema.decodeUnknownOption(ProviderShared.Json)(value), () => value)
|
||||
}
|
||||
|
||||
const comparable = (value: unknown) => {
|
||||
if (!ProviderShared.isRecord(value)) return value
|
||||
if (value.type === "message" && value.role === "assistant")
|
||||
return {
|
||||
role: "assistant",
|
||||
content: value.content,
|
||||
...(value.phase === undefined ? {} : { phase: value.phase }),
|
||||
}
|
||||
if (value.type === "function_call")
|
||||
return {
|
||||
type: value.type,
|
||||
call_id: value.call_id,
|
||||
name: value.name,
|
||||
arguments: json(value.arguments),
|
||||
}
|
||||
if (value.type === "reasoning")
|
||||
return {
|
||||
type: value.type,
|
||||
summary: value.summary,
|
||||
encrypted_content: value.encrypted_content,
|
||||
}
|
||||
return value
|
||||
}
|
||||
|
||||
const invariant = (request: Readonly<Record<string, unknown>>) => {
|
||||
const { type: _type, input: _input, previous_response_id: _previousResponseID, ...rest } = request
|
||||
return rest
|
||||
}
|
||||
|
||||
const incremental = (
|
||||
request: Readonly<Record<string, unknown>>,
|
||||
checkpoint: CheckpointValue,
|
||||
): ReadonlyArray<unknown> | undefined => {
|
||||
const input = request.input
|
||||
const previousInput = checkpoint.request.input
|
||||
if (!Array.isArray(input) || !Array.isArray(previousInput)) return undefined
|
||||
if (canonical(invariant(request)) !== canonical(invariant(checkpoint.request))) return undefined
|
||||
const baseline = [...previousInput, ...checkpoint.output]
|
||||
if (input.length <= baseline.length) return undefined
|
||||
if (!baseline.every((item, index) => canonical(comparable(item)) === canonical(comparable(input[index]))))
|
||||
return undefined
|
||||
return input.slice(baseline.length)
|
||||
}
|
||||
|
||||
const code = (event: OpenResponses.Event) => event.code || event.error?.code || event.response?.error?.code || undefined
|
||||
|
||||
const rejected = (
|
||||
input: DriverInput,
|
||||
observation: Extract<ChannelObservation, { readonly type: "provider-failure" }>,
|
||||
recovery: "retry-full" | "rotate-and-retry-full",
|
||||
): ChannelObservation => ({
|
||||
type: "rejected",
|
||||
recovery,
|
||||
error: new AIError({
|
||||
module: input.id,
|
||||
method: "stream",
|
||||
reason: new TransportReason({
|
||||
message: observation.error.message,
|
||||
transport: "websocket",
|
||||
operation: "read",
|
||||
phase: "receive",
|
||||
delivery: "rejected",
|
||||
recovery,
|
||||
}),
|
||||
}),
|
||||
})
|
||||
|
||||
export const driver = (input: DriverInput): WebSocketChannelDriver => {
|
||||
const { previous_response_id: _previousResponseID, ...request } = input.request
|
||||
let output: unknown[] = []
|
||||
return {
|
||||
create: (checkpoint) =>
|
||||
Effect.sync(() => {
|
||||
output = []
|
||||
const previous = checkpointValue(checkpoint)
|
||||
const delta = previous ? incremental(request, previous) : undefined
|
||||
if (!previous || !delta) return { message: ProviderShared.encodeJson(request), mode: "full" as const }
|
||||
return {
|
||||
message: ProviderShared.encodeJson({ ...request, input: delta, previous_response_id: previous.responseID }),
|
||||
mode: "incremental" as const,
|
||||
}
|
||||
}),
|
||||
observe: (create, frame) =>
|
||||
Effect.gen(function* () {
|
||||
const event = yield* decodeEvent(frame).pipe(
|
||||
Effect.mapError(() => ProviderShared.eventError(input.id, `Invalid ${input.name} WebSocket event`, frame)),
|
||||
)
|
||||
const observation = yield* input.base.observe(create, frame)
|
||||
if (event.type === "response.output_item.done" && event.item) output.push(event.item)
|
||||
if (observation.type === "provider-failure") {
|
||||
const rejection = code(event)
|
||||
if (rejection === "previous_response_not_found") return rejected(input, observation, "retry-full")
|
||||
if (rejection === "websocket_connection_limit_reached")
|
||||
return rejected(input, observation, "rotate-and-retry-full")
|
||||
}
|
||||
if (observation.type !== "completed") return observation
|
||||
const responseID = event.response?.id
|
||||
if (!responseID || responseID.trim().length === 0) return observation
|
||||
return {
|
||||
...observation,
|
||||
checkpoint: {
|
||||
protocol: PROTOCOL,
|
||||
value: { version: VERSION, responseID, request, output: output.slice() } satisfies CheckpointValue,
|
||||
},
|
||||
}
|
||||
}),
|
||||
}
|
||||
}
|
||||
|
||||
export const OpenAIResponsesChannel = { driver } as const
|
||||
@@ -1,266 +0,0 @@
|
||||
import { Effect, Encoding, Schema } from "effect"
|
||||
import { Headers } from "effect/unstable/http"
|
||||
import { Route } from "../route/client.js"
|
||||
import { Auth } from "../route/auth.js"
|
||||
import { Endpoint } from "../route/endpoint.js"
|
||||
import { Protocol } from "../route/protocol.js"
|
||||
import { HttpTransport } from "../route/transport/index.js"
|
||||
import { LLMEvent, LLMRequest, type JsonSchema, type ToolDefinition } from "../schema/index.js"
|
||||
import { OpenResponses } from "./open-responses.js"
|
||||
import { optionalArray, ProviderShared } from "./shared.js"
|
||||
import { Lifecycle } from "./utils/lifecycle.js"
|
||||
import { OpenAIImage } from "./utils/openai-image.js"
|
||||
import { ToolSchemaProjection } from "./utils/tool-schema.js"
|
||||
import { OpenResponsesChannel } from "./open-responses-channel.js"
|
||||
import { OpenAIResponsesChannel } from "./openai-responses-channel.js"
|
||||
|
||||
const ADAPTER = "openai-responses"
|
||||
const NAME = "OpenAI Responses"
|
||||
const WEBSOCKET_PROTOCOL_HEADER = "responses_websockets=2026-02-06"
|
||||
const WEBSOCKET_ROTATE_AFTER_MS = 55 * 60 * 1000
|
||||
export const DEFAULT_BASE_URL = "https://api.openai.com/v1"
|
||||
export const PATH = OpenResponses.PATH
|
||||
|
||||
const OpenAIResponsesImageGenerationTool = Schema.Struct({
|
||||
type: Schema.tag("image_generation"),
|
||||
action: Schema.optional(Schema.Literals(["auto", "generate", "edit"])),
|
||||
background: Schema.optional(Schema.Literals(["auto", "opaque", "transparent"])),
|
||||
input_fidelity: Schema.optional(Schema.Literals(["low", "high"])),
|
||||
output_compression: Schema.optional(Schema.Int.check(Schema.isBetween({ minimum: 0, maximum: 100 }))),
|
||||
output_format: Schema.optional(Schema.Literals(["png", "jpeg", "webp"])),
|
||||
partial_images: Schema.optional(Schema.Int.check(Schema.isGreaterThanOrEqualTo(0))),
|
||||
quality: Schema.optional(Schema.Literals(["auto", "low", "medium", "high"])),
|
||||
size: Schema.optional(OpenAIImage.Size),
|
||||
})
|
||||
|
||||
const OpenAIResponsesTools = Schema.Union([OpenResponses.Tool, OpenAIResponsesImageGenerationTool])
|
||||
|
||||
const OpenAIResponsesToolChoice = Schema.Union([
|
||||
OpenResponses.ToolChoice,
|
||||
Schema.Struct({ type: Schema.tag("image_generation") }),
|
||||
])
|
||||
|
||||
const OpenAIResponsesInputItem = Schema.Union([
|
||||
Schema.Struct({
|
||||
role: Schema.tag("assistant"),
|
||||
content: Schema.Array(Schema.Struct({ type: Schema.tag("output_text"), text: Schema.String })),
|
||||
phase: Schema.optionalKey(Schema.NullOr(OpenResponses.MessagePhase)),
|
||||
}),
|
||||
OpenResponses.InputItem,
|
||||
])
|
||||
|
||||
const OpenAIResponsesCoreFields = {
|
||||
...OpenResponses.coreFields,
|
||||
input: Schema.Array(OpenAIResponsesInputItem),
|
||||
tools: optionalArray(OpenAIResponsesTools),
|
||||
tool_choice: Schema.optional(OpenAIResponsesToolChoice),
|
||||
}
|
||||
|
||||
const OpenAIResponsesBody = Schema.Struct({
|
||||
...OpenAIResponsesCoreFields,
|
||||
stream: Schema.Literal(true),
|
||||
})
|
||||
export type OpenAIResponsesBody = Schema.Schema.Type<typeof OpenAIResponsesBody>
|
||||
|
||||
const extension = {
|
||||
id: ADAPTER,
|
||||
name: NAME,
|
||||
messagePhase: (value: unknown) => (value === null ? null : undefined),
|
||||
lowerMedia: ({ part, media, request }) => {
|
||||
if (request.model.provider !== "xai" || media.mime !== "application/pdf") return undefined
|
||||
return {
|
||||
type: "input_file",
|
||||
filename: part.filename ?? "document.pdf",
|
||||
file_data: media.base64,
|
||||
mime_type: media.mime,
|
||||
}
|
||||
},
|
||||
} satisfies OpenResponses.Extension
|
||||
|
||||
const nativeImageToolInput = (tool: ToolDefinition) => {
|
||||
const native = tool.native?.openai
|
||||
return ProviderShared.isRecord(native) && native.type === "image_generation" ? native : undefined
|
||||
}
|
||||
|
||||
const nativeImageTool = (tool: ToolDefinition) => {
|
||||
const native = nativeImageToolInput(tool)
|
||||
return Schema.is(OpenAIResponsesImageGenerationTool)(native) ? native : undefined
|
||||
}
|
||||
|
||||
const lowerTool = Effect.fn("OpenAIResponses.lowerTool")(function* (tool: ToolDefinition, inputSchema: JsonSchema) {
|
||||
const native = nativeImageToolInput(tool)
|
||||
if (native !== undefined) {
|
||||
if (Schema.is(OpenAIResponsesImageGenerationTool)(native)) return native
|
||||
return yield* ProviderShared.invalidRequest("OpenAI Responses image generation tool options are invalid")
|
||||
}
|
||||
return yield* OpenResponses.lowerTool(NAME, tool, inputSchema)
|
||||
})
|
||||
|
||||
const lowerToolChoice = (toolChoice: NonNullable<LLMRequest["toolChoice"]>, tools: ReadonlyArray<ToolDefinition>) =>
|
||||
ProviderShared.matchToolChoice(NAME, toolChoice, {
|
||||
auto: () => "auto" as const,
|
||||
none: () => "none" as const,
|
||||
required: () => "required" as const,
|
||||
tool: (name) =>
|
||||
tools.some((tool) => tool.name === name && nativeImageTool(tool) !== undefined)
|
||||
? ({ type: "image_generation" } as const)
|
||||
: { type: "function" as const, name },
|
||||
})
|
||||
|
||||
const fromRequest = Effect.fn("OpenAIResponses.fromRequest")(function* (request: LLMRequest) {
|
||||
const body = yield* OpenResponses.fromRequestWithExtension(
|
||||
LLMRequest.update(request, { tools: [], toolChoice: undefined }),
|
||||
extension,
|
||||
)
|
||||
const toolSchemaCompatibility = request.model.compatibility?.toolSchema
|
||||
return {
|
||||
...body,
|
||||
tools:
|
||||
request.tools.length === 0
|
||||
? undefined
|
||||
: yield* Effect.forEach(request.tools, (tool) =>
|
||||
lowerTool(tool, ToolSchemaProjection.modelCompatibility(tool.inputSchema, toolSchemaCompatibility)),
|
||||
),
|
||||
tool_choice: request.toolChoice ? yield* lowerToolChoice(request.toolChoice, request.tools) : undefined,
|
||||
} satisfies OpenAIResponsesBody
|
||||
})
|
||||
|
||||
type HostedToolData = OpenResponses.StreamItem & {
|
||||
readonly id: string
|
||||
readonly status?: string
|
||||
readonly action?: unknown
|
||||
readonly queries?: unknown
|
||||
readonly results?: unknown
|
||||
readonly code?: string
|
||||
readonly container_id?: string
|
||||
readonly outputs?: unknown
|
||||
readonly server_label?: string
|
||||
readonly output?: unknown
|
||||
readonly result?: string
|
||||
readonly output_format?: "png" | "jpeg" | "webp"
|
||||
readonly error?: unknown
|
||||
}
|
||||
|
||||
const HOSTED_TOOLS = {
|
||||
web_search_call: { name: "web_search", input: (item) => item.action ?? {} },
|
||||
web_search_preview_call: { name: "web_search_preview", input: (item) => item.action ?? {} },
|
||||
file_search_call: { name: "file_search", input: (item) => ({ queries: item.queries ?? [] }) },
|
||||
code_interpreter_call: {
|
||||
name: "code_interpreter",
|
||||
input: (item) => ({ code: item.code, container_id: item.container_id }),
|
||||
},
|
||||
computer_use_call: { name: "computer_use", input: (item) => item.action ?? {} },
|
||||
image_generation_call: { name: "image_generation", input: () => ({}) },
|
||||
mcp_call: {
|
||||
name: "mcp",
|
||||
input: (item) => ({ server_label: item.server_label, name: item.name, arguments: item.arguments }),
|
||||
},
|
||||
local_shell_call: { name: "local_shell", input: (item) => item.action ?? {} },
|
||||
} as const satisfies Record<string, { readonly name: string; readonly input: (item: HostedToolData) => unknown }>
|
||||
|
||||
type HostedToolType = keyof typeof HOSTED_TOOLS
|
||||
type HostedToolItem = HostedToolData & { readonly type: HostedToolType }
|
||||
|
||||
const isHostedToolItem = (item: OpenResponses.StreamItem): item is HostedToolItem =>
|
||||
item.type in HOSTED_TOOLS && typeof item.id === "string" && item.id.length > 0
|
||||
|
||||
const hostedToolResult = Effect.fn("OpenAIResponses.hostedToolResult")(function* (item: HostedToolItem) {
|
||||
const isError = item.error !== undefined && item.error !== null
|
||||
if (item.type === "image_generation_call" && item.result) {
|
||||
yield* Effect.fromResult(Encoding.decodeBase64(item.result)).pipe(
|
||||
Effect.mapError(() => ProviderShared.eventError(ADAPTER, "OpenAI Responses returned invalid image base64")),
|
||||
)
|
||||
const format = item.output_format ?? "png"
|
||||
return {
|
||||
type: "content" as const,
|
||||
value: [
|
||||
{
|
||||
type: "file" as const,
|
||||
uri: `data:image/${format};base64,${item.result}`,
|
||||
mime: `image/${format}`,
|
||||
},
|
||||
],
|
||||
}
|
||||
}
|
||||
return isError ? { type: "error" as const, value: item.error } : { type: "json" as const, value: item }
|
||||
})
|
||||
|
||||
const onHostedToolDone = Effect.fn("OpenAIResponses.onHostedToolDone")(function* (
|
||||
state: OpenResponses.ParserState,
|
||||
item: HostedToolItem,
|
||||
) {
|
||||
const tool = HOSTED_TOOLS[item.type]
|
||||
const providerMetadata = OpenResponses.providerMetadata(state, { itemId: item.id })
|
||||
const events: LLMEvent[] = []
|
||||
const lifecycle = Lifecycle.stepStart(state.lifecycle, events)
|
||||
events.push(
|
||||
LLMEvent.toolCall({
|
||||
id: item.id,
|
||||
name: tool.name,
|
||||
input: tool.input(item),
|
||||
providerExecuted: true,
|
||||
providerMetadata,
|
||||
}),
|
||||
LLMEvent.toolResult({
|
||||
id: item.id,
|
||||
name: tool.name,
|
||||
result: yield* hostedToolResult(item),
|
||||
providerExecuted: true,
|
||||
providerMetadata,
|
||||
}),
|
||||
)
|
||||
return [{ ...state, lifecycle }, events] satisfies OpenResponses.StepResult
|
||||
})
|
||||
|
||||
const step = (state: OpenResponses.ParserState, event: OpenResponses.Event) => {
|
||||
if (event.type === "response.reasoning_text.delta" || event.type === "response.reasoning_summary.delta")
|
||||
return event.item_id
|
||||
? Effect.succeed(OpenResponses.onReasoningDelta(state, event, event.item_id))
|
||||
: ProviderShared.eventError(ADAPTER, `${event.type} is missing item_id`)
|
||||
if (event.type === "response.reasoning_text.done" || event.type === "response.reasoning_summary.done")
|
||||
return event.item_id
|
||||
? Effect.succeed(OpenResponses.onReasoningDone(state, event))
|
||||
: ProviderShared.eventError(ADAPTER, `${event.type} is missing item_id`)
|
||||
if (event.type === "response.output_item.done" && event.item && isHostedToolItem(event.item))
|
||||
return onHostedToolDone(state, event.item)
|
||||
return OpenResponses.step(state, event)
|
||||
}
|
||||
|
||||
export const protocol = Protocol.make({
|
||||
id: ADAPTER,
|
||||
body: {
|
||||
schema: OpenAIResponsesBody,
|
||||
from: fromRequest,
|
||||
},
|
||||
stream: {
|
||||
event: OpenResponses.protocol.stream.event,
|
||||
initial: (request) => OpenResponses.initial(request, extension),
|
||||
step,
|
||||
terminal: OpenResponses.terminal,
|
||||
},
|
||||
})
|
||||
|
||||
const endpoint = Endpoint.path<OpenAIResponsesBody>(PATH, { baseURL: DEFAULT_BASE_URL })
|
||||
const auth = Auth.none
|
||||
|
||||
export const httpTransport = HttpTransport.sseJson.with<OpenAIResponsesBody>()
|
||||
export const transport = OpenResponsesChannel.transport<OpenAIResponsesBody>({
|
||||
id: ADAPTER,
|
||||
name: NAME,
|
||||
rotateAfterMs: WEBSOCKET_ROTATE_AFTER_MS,
|
||||
headers: (headers) => Headers.set(headers, "openai-beta", headers["openai-beta"] ?? WEBSOCKET_PROTOCOL_HEADER),
|
||||
driver: (input) => OpenAIResponsesChannel.driver({ id: ADAPTER, name: NAME, ...input }),
|
||||
})
|
||||
|
||||
export const route = Route.make({
|
||||
id: ADAPTER,
|
||||
provider: "openai",
|
||||
providerMetadataKey: "openai",
|
||||
protocol,
|
||||
endpoint,
|
||||
auth,
|
||||
transport,
|
||||
defaults: { providerOptions: { openai: { store: false } } },
|
||||
})
|
||||
|
||||
export * as OpenAIResponses from "./openai-responses.js"
|
||||
@@ -1,327 +0,0 @@
|
||||
import { Buffer } from "node:buffer"
|
||||
import { Tool } from "@opencode-ai/schema/tool"
|
||||
import { Effect, Schema, Stream } from "effect"
|
||||
import * as Sse from "effect/unstable/encoding/Sse"
|
||||
import { Headers, HttpClientRequest } from "effect/unstable/http"
|
||||
import {
|
||||
InvalidProviderOutputReason,
|
||||
InvalidRequestReason,
|
||||
AIError,
|
||||
type ContentPart,
|
||||
type LLMRequest,
|
||||
type MediaPart,
|
||||
type TextPart,
|
||||
type ToolResultPart,
|
||||
} from "../schema/index.js"
|
||||
import { isRecord } from "../utils/record.js"
|
||||
export { isRecord }
|
||||
|
||||
export const Json = Schema.fromJsonString(Schema.Unknown)
|
||||
export const decodeJson = Schema.decodeUnknownSync(Json)
|
||||
export const encodeJson = Schema.encodeSync(Json)
|
||||
const isJson = Schema.is(Schema.Json)
|
||||
export const JsonObject = Schema.Record(Schema.String, Schema.Unknown)
|
||||
export const optionalArray = <const S extends Schema.Top>(schema: S) => Schema.optional(Schema.Array(schema))
|
||||
export const optionalNull = <const S extends Schema.Top>(schema: S) => Schema.optional(Schema.NullOr(schema))
|
||||
|
||||
/**
|
||||
* Streaming tool-call accumulator. Adapters that build a tool call across
|
||||
* multiple `tool-input-delta` chunks store the partial JSON input string here
|
||||
* and finalize it with `parseToolInput` once the call completes.
|
||||
*/
|
||||
export interface ToolAccumulator {
|
||||
readonly id: string
|
||||
readonly name: string
|
||||
readonly input: string
|
||||
}
|
||||
|
||||
/**
|
||||
* `Usage.totalTokens` policy shared by every route. Honors a provider-
|
||||
* supplied total; otherwise falls back to `inputTokens + outputTokens` only
|
||||
* when at least one is defined. Returns `undefined` when neither input nor
|
||||
* output is known so routes don't publish a misleading `0`.
|
||||
*
|
||||
* Under the additive `AI.Usage` contract, `inputTokens` and `outputTokens`
|
||||
* are the non-cached input and visible output only. The provider-supplied
|
||||
* `total` is the source of truth when present; the computed fallback
|
||||
* under-counts cache and reasoning by design and exists mainly so
|
||||
* Anthropic-style providers (which don't surface a total) still get a
|
||||
* sensible aggregate on the input + output axes.
|
||||
*/
|
||||
export const totalTokens = (
|
||||
inputTokens: number | undefined,
|
||||
outputTokens: number | undefined,
|
||||
total: number | undefined,
|
||||
) => {
|
||||
if (total !== undefined) return total
|
||||
if (inputTokens === undefined && outputTokens === undefined) return undefined
|
||||
return (inputTokens ?? 0) + (outputTokens ?? 0)
|
||||
}
|
||||
|
||||
/**
|
||||
* Subtract `subtrahend` from `total`, clamping to zero if the provider
|
||||
* reports a non-sensical breakdown (e.g. `cached_tokens > prompt_tokens`).
|
||||
* Used by protocol mappers when deriving a non-overlapping breakdown field
|
||||
* from a provider's inclusive total — `nonCachedInputTokens` from
|
||||
* `inputTokens - cacheReadInputTokens - cacheWriteInputTokens`.
|
||||
*
|
||||
* If `total` is `undefined`, returns `undefined` (we don't fabricate
|
||||
* counts). If `subtrahend` is `undefined`, returns `total` unchanged. The
|
||||
* provider-native breakdown stays available on `Usage.native` for debugging.
|
||||
*/
|
||||
export const subtractTokens = (total: number | undefined, subtrahend: number | undefined): number | undefined => {
|
||||
if (total === undefined) return undefined
|
||||
if (subtrahend === undefined) return total
|
||||
return Math.max(0, total - subtrahend)
|
||||
}
|
||||
|
||||
/**
|
||||
* Sum a list of optional token counts, returning `undefined` only when
|
||||
* every value is `undefined` (so we don't fabricate a `0`). Used by
|
||||
* protocol mappers to derive the inclusive `inputTokens` total from a
|
||||
* provider that natively reports a non-overlapping breakdown
|
||||
* (e.g. Anthropic, whose `input_tokens` is already non-cached only).
|
||||
*/
|
||||
export const sumTokens = (...values: ReadonlyArray<number | undefined>): number | undefined => {
|
||||
if (values.every((value) => value === undefined)) return undefined
|
||||
return values.reduce((acc: number, value) => acc + (value ?? 0), 0)
|
||||
}
|
||||
|
||||
export const eventError = (route: string, message: string, raw?: string) =>
|
||||
new AIError({
|
||||
module: "ProviderShared",
|
||||
method: "stream",
|
||||
reason: new InvalidProviderOutputReason({ route, message, raw }),
|
||||
})
|
||||
|
||||
export const parseJson = (route: string, input: string, message: string) =>
|
||||
Effect.try({
|
||||
try: () => decodeJson(input),
|
||||
catch: () => eventError(route, message, input),
|
||||
})
|
||||
|
||||
/**
|
||||
* Join the `text` field of a list of parts with newlines. Used by routes
|
||||
* that flatten system / message content arrays into a single provider string
|
||||
* (OpenAI Chat `system` content, OpenAI Responses `system` content, Gemini
|
||||
* `systemInstruction.parts[].text`).
|
||||
*/
|
||||
export const joinText = (parts: ReadonlyArray<{ readonly text: string }>) => parts.map((part) => part.text).join("\n")
|
||||
|
||||
const escapeSystemUpdateText = (text: string) =>
|
||||
text.replaceAll("&", "&").replaceAll("<", "<").replaceAll(">", ">")
|
||||
|
||||
/**
|
||||
* Stable fallback representation for chronological `Message.system(...)`
|
||||
* updates on routes that do not support that privileged role natively. The
|
||||
* wrapper remains visibly lower-authority user text, preserves the original
|
||||
* temporal position, and XML-escapes content so it cannot close the wrapper.
|
||||
*/
|
||||
export const wrapSystemUpdate = (parts: ReadonlyArray<{ readonly text: string }>) =>
|
||||
`<system-update>\n${escapeSystemUpdateText(joinText(parts))}\n</system-update>`
|
||||
|
||||
/**
|
||||
* Chronological system updates deliberately accept text only. Do not insert
|
||||
* raw retrieved, tool, or web content into privileged updates: keep untrusted
|
||||
* data in ordinary user/tool messages instead.
|
||||
*/
|
||||
export const systemUpdateText = Effect.fn("ProviderShared.systemUpdateText")(function* (
|
||||
route: string,
|
||||
message: LLMRequest["messages"][number],
|
||||
) {
|
||||
const content: TextPart[] = []
|
||||
for (const part of message.content) {
|
||||
if (!supportsContent(part, ["text"])) return yield* unsupportedContent(route, "system", ["text"])
|
||||
content.push(part)
|
||||
}
|
||||
return content
|
||||
})
|
||||
|
||||
/** Lower an unsupported privileged update into visible, in-order user text. */
|
||||
export const wrappedSystemUpdate = Effect.fn("ProviderShared.wrappedSystemUpdate")(function* (
|
||||
route: string,
|
||||
message: LLMRequest["messages"][number],
|
||||
) {
|
||||
const content = yield* systemUpdateText(route, message)
|
||||
return { type: "text" as const, text: wrapSystemUpdate(content), cache: content.at(-1)?.cache }
|
||||
})
|
||||
|
||||
/**
|
||||
* Parse the streamed JSON input of a tool call. Treats an empty string as
|
||||
* `"{}"` — providers occasionally finish a tool call without ever emitting
|
||||
* input deltas (e.g. zero-arg tools). The error message is uniform across
|
||||
* routes: `Invalid JSON input for <route> tool call <name>`.
|
||||
*/
|
||||
export const parseToolInput = (route: string, name: string, raw: string) =>
|
||||
parseJson(route, raw || "{}", `Invalid JSON input for ${route} tool call ${name}`)
|
||||
|
||||
export const IMAGE_MIMES = ["image/png", "image/jpeg", "image/gif", "image/webp"] as const
|
||||
export const VIDEO_MIMES = ["video/mp4", "video/webm", "video/quicktime"] as const
|
||||
export const AUDIO_MIMES = ["audio/wav", "audio/mp3", "audio/aiff", "audio/aac", "audio/ogg", "audio/flac"] as const
|
||||
export const PDF_MIMES = ["application/pdf"] as const
|
||||
export const MEDIA_MIMES = [...IMAGE_MIMES, ...VIDEO_MIMES, ...AUDIO_MIMES, ...PDF_MIMES] as const
|
||||
export const MAX_MEDIA_ENCODED_BYTES = 28 * 1024 * 1024
|
||||
export const MAX_MEDIA_DECODED_BYTES = 20 * 1024 * 1024
|
||||
|
||||
const base64Pattern = /^(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}==|[A-Za-z0-9+/]{3}=)?$/
|
||||
|
||||
export interface ValidatedMedia {
|
||||
readonly mime: string
|
||||
readonly base64: string
|
||||
readonly dataUrl: string
|
||||
readonly bytes: Uint8Array
|
||||
}
|
||||
|
||||
export const validateMedia = Effect.fn("ProviderShared.validateMedia")(function* (
|
||||
route: string,
|
||||
part: MediaPart,
|
||||
supportedMimes: ReadonlySet<string>,
|
||||
) {
|
||||
const mime = part.mediaType.toLowerCase()
|
||||
if (!supportedMimes.has(mime)) return yield* invalidRequest(`${route} does not support media type ${part.mediaType}`)
|
||||
|
||||
let base64: string
|
||||
if (typeof part.data !== "string") {
|
||||
if (part.data.byteLength > MAX_MEDIA_DECODED_BYTES)
|
||||
return yield* invalidRequest(`${route} media exceeds the ${MAX_MEDIA_DECODED_BYTES} byte decoded limit`)
|
||||
base64 = Buffer.from(part.data).toString("base64")
|
||||
} else if (part.data.startsWith("data:")) {
|
||||
const match = /^data:([^;,]+);base64,([A-Za-z0-9+/]*={0,2})$/s.exec(part.data)
|
||||
if (!match) return yield* invalidRequest(`${route} media data URL must contain valid base64`)
|
||||
if (match[1]!.toLowerCase() !== mime)
|
||||
return yield* invalidRequest(`${route} media type ${part.mediaType} does not match data URL type ${match[1]}`)
|
||||
base64 = match[2]!
|
||||
} else {
|
||||
base64 = part.data
|
||||
}
|
||||
|
||||
if (Buffer.byteLength(base64, "utf8") > MAX_MEDIA_ENCODED_BYTES)
|
||||
return yield* invalidRequest(`${route} media exceeds the ${MAX_MEDIA_ENCODED_BYTES} byte encoded limit`)
|
||||
if (!base64 || base64.length % 4 !== 0 || !base64Pattern.test(base64))
|
||||
return yield* invalidRequest(`${route} media must contain valid base64`)
|
||||
const bytes = Buffer.from(base64, "base64")
|
||||
if (bytes.byteLength > MAX_MEDIA_DECODED_BYTES)
|
||||
return yield* invalidRequest(`${route} media exceeds the ${MAX_MEDIA_DECODED_BYTES} byte decoded limit`)
|
||||
if (bytes.toString("base64") !== base64) return yield* invalidRequest(`${route} media must contain canonical base64`)
|
||||
return { mime, base64, dataUrl: `data:${mime};base64,${base64}`, bytes } satisfies ValidatedMedia
|
||||
})
|
||||
|
||||
export const validateToolFile = (route: string, part: Tool.FileContent, supportedMimes: ReadonlySet<string>) =>
|
||||
validateMedia(route, { type: "media", mediaType: part.mime, data: part.uri, filename: part.name }, supportedMimes)
|
||||
|
||||
export const trimBaseUrl = (value: string) => value.replace(/\/+$/, "")
|
||||
|
||||
export const toolResultText = (part: ToolResultPart) => {
|
||||
if (part.result.type === "text") return String(part.result.value)
|
||||
if (part.result.type === "error") {
|
||||
const value = part.result.value
|
||||
const prototype =
|
||||
typeof value === "object" && value !== null && !Array.isArray(value) && Object.getPrototypeOf(value)
|
||||
const structured = Array.isArray(value) || prototype === Object.prototype || prototype === null
|
||||
return structured && isJson(value) ? encodeJson(value) : String(value)
|
||||
}
|
||||
return encodeJson(part.result.value)
|
||||
}
|
||||
|
||||
export const errorText = (error: unknown) => {
|
||||
if (error instanceof Error) return error.message
|
||||
if (typeof error === "string") return error
|
||||
if (typeof error === "number" || typeof error === "boolean" || typeof error === "bigint") return String(error)
|
||||
if (error === null) return "null"
|
||||
if (error === undefined) return "undefined"
|
||||
return "Unknown stream error"
|
||||
}
|
||||
|
||||
/**
|
||||
* `framing` step for Server-Sent Events. Decodes UTF-8, runs the SSE channel
|
||||
* decoder, and drops empty / `[DONE]` keep-alive events so the downstream
|
||||
* `decodeChunk` sees one JSON string per element. The SSE channel emits a
|
||||
* `Retry` control event on its error channel; we drop it here (we don't
|
||||
* implement client-driven retries) so the public error channel stays
|
||||
* `AIError`.
|
||||
*/
|
||||
export const sseFraming = (bytes: Stream.Stream<Uint8Array, AIError>): Stream.Stream<string, AIError> =>
|
||||
bytes.pipe(
|
||||
Stream.decodeText(),
|
||||
Stream.pipeThroughChannel(Sse.decode()),
|
||||
Stream.catchTag("Retry", () => Stream.empty),
|
||||
Stream.filter((event) => event.data.length > 0 && event.data !== "[DONE]"),
|
||||
Stream.map((event) => event.data),
|
||||
)
|
||||
|
||||
/**
|
||||
* Canonical invalid-request constructor. Lift one-line `const invalid =
|
||||
* (message) => invalidRequest(message)` aliases out of every
|
||||
* route so the error constructor lives in one place. If we ever extend
|
||||
* `InvalidRequestReason` with route context or trace metadata, the change
|
||||
* lands here.
|
||||
*/
|
||||
export const invalidRequest = (message: string) =>
|
||||
new AIError({
|
||||
module: "ProviderShared",
|
||||
method: "request",
|
||||
reason: new InvalidRequestReason({ message }),
|
||||
})
|
||||
|
||||
export const matchToolChoice = <Auto, None, Required, Tool>(
|
||||
route: string,
|
||||
toolChoice: NonNullable<LLMRequest["toolChoice"]>,
|
||||
cases: {
|
||||
readonly auto: () => Auto
|
||||
readonly none: () => None
|
||||
readonly required: () => Required
|
||||
readonly tool: (name: string) => Tool
|
||||
},
|
||||
) =>
|
||||
Effect.gen(function* () {
|
||||
if (toolChoice.type === "auto") return cases.auto()
|
||||
if (toolChoice.type === "none") return cases.none()
|
||||
if (toolChoice.type === "required") return cases.required()
|
||||
if (!toolChoice.name) return yield* invalidRequest(`${route} tool choice requires a tool name`)
|
||||
return cases.tool(toolChoice.name)
|
||||
})
|
||||
|
||||
type ContentType = ContentPart["type"]
|
||||
|
||||
const formatContentTypes = (types: ReadonlyArray<ContentType>) => {
|
||||
if (types.length <= 1) return types[0] ?? ""
|
||||
if (types.length === 2) return `${types[0]} and ${types[1]}`
|
||||
return `${types.slice(0, -1).join(", ")}, and ${types.at(-1)}`
|
||||
}
|
||||
|
||||
export const supportsContent = <const Type extends ContentType>(
|
||||
part: ContentPart,
|
||||
types: ReadonlyArray<Type>,
|
||||
): part is Extract<ContentPart, { readonly type: Type }> => (types as ReadonlyArray<ContentType>).includes(part.type)
|
||||
|
||||
export const unsupportedContent = (
|
||||
route: string,
|
||||
role: LLMRequest["messages"][number]["role"],
|
||||
types: ReadonlyArray<ContentType>,
|
||||
) => invalidRequest(`${route} ${role} messages only support ${formatContentTypes(types)} content for now`)
|
||||
|
||||
/**
|
||||
* Build a `validate` step from a Schema decoder. Replaces the per-route
|
||||
* lambda body `(payload) => decode(payload).pipe(Effect.mapError((e) =>
|
||||
* invalid(e.message)))`. Any decode error is translated into
|
||||
* `AIError` carrying the original parse-error message.
|
||||
*/
|
||||
export const validateWith =
|
||||
<A, I, E extends { readonly message: string }>(decode: (input: I) => Effect.Effect<A, E>) =>
|
||||
(payload: I) =>
|
||||
decode(payload).pipe(Effect.mapError((error) => invalidRequest(error.message)))
|
||||
|
||||
/**
|
||||
* Build an HTTP POST with a JSON body. Sets `content-type: application/json`
|
||||
* automatically after caller-supplied headers so routes cannot accidentally
|
||||
* send JSON with a stale content type. The body is passed pre-encoded so
|
||||
* routes can choose between
|
||||
* `Schema.encodeSync(payload)` and `ProviderShared.encodeJson(payload)`.
|
||||
*/
|
||||
export const jsonPost = (input: { readonly url: string; readonly body: string; readonly headers?: Headers.Input }) =>
|
||||
HttpClientRequest.post(input.url).pipe(
|
||||
HttpClientRequest.setHeaders(Headers.set(Headers.fromInput(input.headers), "content-type", "application/json")),
|
||||
HttpClientRequest.bodyText(input.body, "application/json"),
|
||||
)
|
||||
|
||||
export * as ProviderShared from "./shared.js"
|
||||
@@ -1,77 +0,0 @@
|
||||
import { AwsV4Signer } from "aws4fetch"
|
||||
import { Effect } from "effect"
|
||||
import { Headers } from "effect/unstable/http"
|
||||
import { Auth, type AuthInput } from "../../route/auth.js"
|
||||
import { ProviderShared } from "../shared.js"
|
||||
|
||||
/**
|
||||
* AWS credentials for SigV4 signing. Bedrock also supports Bearer API key auth,
|
||||
* which provider facades configure as route auth instead of SigV4. STS-vended
|
||||
* credentials should be refreshed by the consumer (rebuild the model) before
|
||||
* they expire; the route does not refresh.
|
||||
*/
|
||||
export interface Credentials {
|
||||
readonly region: string
|
||||
readonly accessKeyId: string
|
||||
readonly secretAccessKey: string
|
||||
readonly sessionToken?: string
|
||||
}
|
||||
|
||||
const signRequest = (input: {
|
||||
readonly url: string
|
||||
readonly body: string
|
||||
readonly headers: Headers.Headers
|
||||
readonly credentials: Credentials
|
||||
readonly service: string
|
||||
readonly name: string
|
||||
}) =>
|
||||
Effect.tryPromise({
|
||||
try: async () => {
|
||||
const signed = await new AwsV4Signer({
|
||||
url: input.url,
|
||||
method: "POST",
|
||||
headers: Object.entries(input.headers),
|
||||
body: input.body,
|
||||
region: input.credentials.region,
|
||||
accessKeyId: input.credentials.accessKeyId,
|
||||
secretAccessKey: input.credentials.secretAccessKey,
|
||||
sessionToken: input.credentials.sessionToken,
|
||||
service: input.service,
|
||||
}).sign()
|
||||
return Object.fromEntries(signed.headers.entries())
|
||||
},
|
||||
catch: (error) =>
|
||||
ProviderShared.invalidRequest(
|
||||
`${input.name} SigV4 signing failed: ${error instanceof Error ? error.message : String(error)}`,
|
||||
),
|
||||
})
|
||||
|
||||
/** Sign the exact JSON bytes with SigV4 using credentials configured on the route. */
|
||||
export const sigV4 = (
|
||||
credentials: Credentials | undefined,
|
||||
options: { readonly service?: string; readonly name?: string } = {},
|
||||
) =>
|
||||
Auth.custom((input: AuthInput) => {
|
||||
return Effect.gen(function* () {
|
||||
if (!credentials) {
|
||||
return yield* ProviderShared.invalidRequest(
|
||||
`${options.name ?? "Bedrock Converse"} requires either route bearer auth or AWS credentials configured on the route`,
|
||||
)
|
||||
}
|
||||
const headersForSigning = Headers.set(input.headers, "content-type", "application/json")
|
||||
const signed = yield* signRequest({
|
||||
url: input.url,
|
||||
body: input.body,
|
||||
headers: headersForSigning,
|
||||
credentials,
|
||||
service: options.service ?? "bedrock",
|
||||
name: options.name ?? "Bedrock Converse",
|
||||
})
|
||||
return Headers.setAll(headersForSigning, signed)
|
||||
})
|
||||
})
|
||||
|
||||
/** Bedrock route auth defaults to SigV4 and expects credentials from route configuration. */
|
||||
export const auth = sigV4(undefined)
|
||||
|
||||
export * as BedrockAuth from "./bedrock-auth.js"
|
||||
@@ -1,119 +0,0 @@
|
||||
import { isRecord } from "../../utils/record.js"
|
||||
|
||||
// Gemini accepts a JSON Schema-like dialect for tool parameters, but rejects a
|
||||
// handful of common JSON Schema shapes. Keep this projection isolated so the
|
||||
// Gemini protocol file still reads like the other protocol modules.
|
||||
const SCHEMA_INTENT_KEYS = [
|
||||
"type",
|
||||
"properties",
|
||||
"items",
|
||||
"prefixItems",
|
||||
"enum",
|
||||
"const",
|
||||
"$ref",
|
||||
"additionalProperties",
|
||||
"patternProperties",
|
||||
"required",
|
||||
"not",
|
||||
"if",
|
||||
"then",
|
||||
"else",
|
||||
]
|
||||
|
||||
const hasCombiner = (schema: unknown) =>
|
||||
isRecord(schema) && (Array.isArray(schema.anyOf) || Array.isArray(schema.oneOf) || Array.isArray(schema.allOf))
|
||||
|
||||
const hasSchemaIntent = (schema: unknown) =>
|
||||
isRecord(schema) && (hasCombiner(schema) || SCHEMA_INTENT_KEYS.some((key) => key in schema))
|
||||
|
||||
const sanitizeNode = (schema: unknown): unknown => {
|
||||
if (!isRecord(schema)) return Array.isArray(schema) ? schema.map(sanitizeNode) : schema
|
||||
|
||||
const result: Record<string, unknown> = Object.fromEntries(
|
||||
Object.entries(schema).map(([key, value]) => [
|
||||
key,
|
||||
key === "enum" && Array.isArray(value) ? value.map(String) : sanitizeNode(value),
|
||||
]),
|
||||
)
|
||||
|
||||
if (Array.isArray(result.enum) && (result.type === "integer" || result.type === "number")) result.type = "string"
|
||||
|
||||
const properties = result.properties
|
||||
if (result.type === "object" && isRecord(properties) && Array.isArray(result.required)) {
|
||||
result.required = result.required.filter((field) => typeof field === "string" && field in properties)
|
||||
}
|
||||
|
||||
if (result.type === "array" && !hasCombiner(result)) {
|
||||
result.items = result.items ?? {}
|
||||
if (isRecord(result.items) && !hasSchemaIntent(result.items)) result.items = { ...result.items, type: "string" }
|
||||
}
|
||||
|
||||
if (typeof result.type === "string" && result.type !== "object" && !hasCombiner(result)) {
|
||||
delete result.properties
|
||||
delete result.required
|
||||
}
|
||||
|
||||
return result
|
||||
}
|
||||
|
||||
const emptyObjectSchema = (schema: Record<string, unknown>) =>
|
||||
schema.type === "object" &&
|
||||
(!isRecord(schema.properties) || Object.keys(schema.properties).length === 0) &&
|
||||
!schema.additionalProperties
|
||||
|
||||
const projectNode = (schema: unknown, nested = false): Record<string, unknown> | undefined => {
|
||||
if (!isRecord(schema)) return undefined
|
||||
if (!nested && emptyObjectSchema(schema)) return undefined
|
||||
const types = Array.isArray(schema.type) ? schema.type.filter((type) => type !== "null") : undefined
|
||||
const anyOf = Array.isArray(schema.anyOf) ? schema.anyOf : undefined
|
||||
const hasNullAnyOf = anyOf?.some((item) => isRecord(item) && item.type === "null") ?? false
|
||||
const anyOfTypes = hasNullAnyOf ? anyOf?.filter((item) => !isRecord(item) || item.type !== "null") : anyOf
|
||||
const flattenedAnyOf = hasNullAnyOf && anyOfTypes?.length === 1 ? projectNode(anyOfTypes[0], true) : undefined
|
||||
const result = Object.fromEntries(
|
||||
[
|
||||
["description", schema.description],
|
||||
["required", schema.required],
|
||||
["format", schema.format],
|
||||
["type", types ? (types.length === 0 ? "null" : undefined) : schema.type],
|
||||
[
|
||||
"nullable",
|
||||
(Array.isArray(schema.type) && schema.type.includes("null") && types && types.length > 0) || hasNullAnyOf
|
||||
? true
|
||||
: undefined,
|
||||
],
|
||||
["enum", schema.const !== undefined ? [schema.const] : schema.enum],
|
||||
[
|
||||
"properties",
|
||||
isRecord(schema.properties)
|
||||
? Object.fromEntries(Object.entries(schema.properties).map(([key, value]) => [key, projectNode(value, true)]))
|
||||
: undefined,
|
||||
],
|
||||
[
|
||||
"items",
|
||||
Array.isArray(schema.items)
|
||||
? schema.items.map((item) => projectNode(item, true))
|
||||
: schema.items === undefined
|
||||
? undefined
|
||||
: projectNode(schema.items, true),
|
||||
],
|
||||
["allOf", Array.isArray(schema.allOf) ? schema.allOf.map((item) => projectNode(item, true)) : undefined],
|
||||
[
|
||||
"anyOf",
|
||||
anyOfTypes
|
||||
? hasNullAnyOf && anyOfTypes.length === 1
|
||||
? undefined
|
||||
: anyOfTypes.map((item) => projectNode(item, true))
|
||||
: types && types.length > 0
|
||||
? types.map((type) => ({ type }))
|
||||
: undefined,
|
||||
],
|
||||
["oneOf", Array.isArray(schema.oneOf) ? schema.oneOf.map((item) => projectNode(item, true)) : undefined],
|
||||
["minLength", schema.minLength],
|
||||
].filter((entry) => entry[1] !== undefined),
|
||||
)
|
||||
return flattenedAnyOf ? { ...result, ...flattenedAnyOf } : result
|
||||
}
|
||||
|
||||
export const convert = (schema: unknown) => projectNode(sanitizeNode(schema))
|
||||
|
||||
export * as GeminiToolSchema from "./gemini-tool-schema.js"
|
||||
@@ -1,34 +0,0 @@
|
||||
import { Effect, Encoding } from "effect"
|
||||
import type { ImageInput } from "../../image.js"
|
||||
import { InvalidRequestReason, AIError } from "../../schema/index.js"
|
||||
|
||||
const invalid = (module: string, message: string) =>
|
||||
new AIError({
|
||||
module,
|
||||
method: "generate",
|
||||
reason: new InvalidRequestReason({ message }),
|
||||
})
|
||||
|
||||
export const dataUrl = (input: Extract<ImageInput, { readonly type: "bytes" }>) =>
|
||||
`data:${input.mediaType};base64,${Encoding.encodeBase64(input.data)}`
|
||||
|
||||
export const decodeDataUrl = (
|
||||
url: string,
|
||||
module: string,
|
||||
): Effect.Effect<{ readonly mediaType: string; readonly data: Uint8Array } | undefined, AIError> => {
|
||||
if (!url.startsWith("data:")) return Effect.succeed(undefined)
|
||||
const match = /^data:([^;,]+);base64,(.*)$/s.exec(url)
|
||||
if (!match) return Effect.fail(invalid(module, "Image data URLs must contain a MIME type and base64 data"))
|
||||
return Effect.fromResult(Encoding.decodeBase64(match[2])).pipe(
|
||||
Effect.mapError(() => invalid(module, "Image data URL contains invalid base64 data")),
|
||||
Effect.map((data) => ({ mediaType: match[1], data })),
|
||||
)
|
||||
}
|
||||
|
||||
export const invalidImageInput = invalid
|
||||
|
||||
export const ImageInputs = {
|
||||
dataUrl,
|
||||
decodeDataUrl,
|
||||
invalid: invalidImageInput,
|
||||
} as const
|
||||
@@ -1,105 +0,0 @@
|
||||
import { LLMEvent, type FinishReasonDetails, type ProviderMetadata, type Usage } from "../../schema/index.js"
|
||||
|
||||
export interface State {
|
||||
readonly stepStarted: boolean
|
||||
readonly text: ReadonlySet<string>
|
||||
readonly reasoning: ReadonlySet<string>
|
||||
}
|
||||
|
||||
export const initial = (): State => ({ stepStarted: false, text: new Set(), reasoning: new Set() })
|
||||
|
||||
export const stepStart = (state: State, events: LLMEvent[]): State => {
|
||||
if (state.stepStarted) return state
|
||||
events.push(LLMEvent.stepStart({ index: 0 }))
|
||||
return { ...state, stepStarted: true }
|
||||
}
|
||||
|
||||
export const textStart = (state: State, events: LLMEvent[], id: string, providerMetadata?: ProviderMetadata): State => {
|
||||
if (state.text.has(id)) return state
|
||||
const stepped = stepStart(state, events)
|
||||
events.push(LLMEvent.textStart({ id, providerMetadata }))
|
||||
return { ...stepped, text: new Set([...stepped.text, id]) }
|
||||
}
|
||||
|
||||
export const textDelta = (state: State, events: LLMEvent[], id: string, text: string): State => {
|
||||
const started = textStart(state, events, id)
|
||||
events.push(LLMEvent.textDelta({ id, text }))
|
||||
return started
|
||||
}
|
||||
|
||||
export const reasoningStart = (
|
||||
state: State,
|
||||
events: LLMEvent[],
|
||||
id: string,
|
||||
providerMetadata?: ProviderMetadata,
|
||||
): State => {
|
||||
if (state.reasoning.has(id)) return state
|
||||
const stepped = stepStart(state, events)
|
||||
events.push(LLMEvent.reasoningStart({ id, providerMetadata }))
|
||||
return { ...stepped, reasoning: new Set([...stepped.reasoning, id]) }
|
||||
}
|
||||
|
||||
export const reasoningDelta = (
|
||||
state: State,
|
||||
events: LLMEvent[],
|
||||
id: string,
|
||||
text: string,
|
||||
providerMetadata?: ProviderMetadata,
|
||||
): State => {
|
||||
const started = reasoningStart(state, events, id, providerMetadata)
|
||||
events.push(LLMEvent.reasoningDelta({ id, text, providerMetadata }))
|
||||
return started
|
||||
}
|
||||
|
||||
export const reasoningEnd = (
|
||||
state: State,
|
||||
events: LLMEvent[],
|
||||
id: string,
|
||||
providerMetadata?: ProviderMetadata,
|
||||
): State => {
|
||||
if (!state.reasoning.has(id)) return state
|
||||
const stepped = stepStart(state, events)
|
||||
events.push(LLMEvent.reasoningEnd({ id, providerMetadata }))
|
||||
const reasoning = new Set(stepped.reasoning)
|
||||
reasoning.delete(id)
|
||||
return { ...stepped, reasoning }
|
||||
}
|
||||
|
||||
export const textEnd = (state: State, events: LLMEvent[], id: string, providerMetadata?: ProviderMetadata): State => {
|
||||
if (!state.text.has(id)) return state
|
||||
const stepped = stepStart(state, events)
|
||||
events.push(LLMEvent.textEnd({ id, providerMetadata }))
|
||||
const text = new Set(stepped.text)
|
||||
text.delete(id)
|
||||
return { ...stepped, text }
|
||||
}
|
||||
|
||||
const closeOpenBlocks = (state: State, events: LLMEvent[]): State => {
|
||||
for (const id of state.reasoning) events.push(LLMEvent.reasoningEnd({ id }))
|
||||
for (const id of state.text) events.push(LLMEvent.textEnd({ id }))
|
||||
return { ...state, text: new Set(), reasoning: new Set() }
|
||||
}
|
||||
|
||||
export const finish = (
|
||||
state: State,
|
||||
events: LLMEvent[],
|
||||
input: {
|
||||
readonly reason: FinishReasonDetails
|
||||
readonly usage?: Usage
|
||||
readonly providerMetadata?: ProviderMetadata
|
||||
},
|
||||
): State => {
|
||||
const stepped = closeOpenBlocks(stepStart(state, events), events)
|
||||
events.push(
|
||||
LLMEvent.stepFinish({
|
||||
index: 0,
|
||||
reason: input.reason,
|
||||
usage: input.usage,
|
||||
providerMetadata: input.providerMetadata,
|
||||
}),
|
||||
LLMEvent.finish(input),
|
||||
)
|
||||
return { ...stepped, stepStarted: false }
|
||||
}
|
||||
|
||||
export * as Lifecycle from "./lifecycle.js"
|
||||
@@ -1,63 +0,0 @@
|
||||
import { Schema } from "effect"
|
||||
import { TextVerbosity, type LLMRequest } from "../../schema/index.js"
|
||||
|
||||
export const ResponseIncludables = [
|
||||
"file_search_call.results",
|
||||
"web_search_call.results",
|
||||
"web_search_call.action.sources",
|
||||
"message.input_image.image_url",
|
||||
"computer_call_output.output.image_url",
|
||||
"code_interpreter_call.outputs",
|
||||
"reasoning.encrypted_content",
|
||||
"message.output_text.logprobs",
|
||||
] as const
|
||||
export type ResponseIncludable = (typeof ResponseIncludables)[number]
|
||||
|
||||
export const ServiceTiers = ["auto", "default", "flex", "priority"] as const
|
||||
export type ServiceTier = (typeof ServiceTiers)[number]
|
||||
|
||||
const TEXT_VERBOSITY = new Set<string>(["low", "medium", "high"])
|
||||
const INCLUDABLES = new Set<string>(ResponseIncludables)
|
||||
const SERVICE_TIERS = new Set<string>(ServiceTiers)
|
||||
|
||||
const isTextVerbosity = (value: unknown): value is Schema.Schema.Type<typeof TextVerbosity> =>
|
||||
typeof value === "string" && TEXT_VERBOSITY.has(value)
|
||||
|
||||
const isServiceTier = (value: unknown): value is ServiceTier => typeof value === "string" && SERVICE_TIERS.has(value)
|
||||
|
||||
export const ReasoningEffort = Schema.String
|
||||
export const TextVerbositySchema = TextVerbosity
|
||||
export const ResponseIncludableSchema = Schema.Literals(ResponseIncludables)
|
||||
export const ServiceTierSchema = Schema.Literals(ServiceTiers)
|
||||
|
||||
export interface Resolved {
|
||||
readonly instructions?: string
|
||||
readonly store?: boolean
|
||||
readonly reasoningEffort?: string
|
||||
readonly reasoningSummary?: "auto" | "concise" | "detailed"
|
||||
readonly include?: ReadonlyArray<ResponseIncludable>
|
||||
readonly textVerbosity?: Schema.Schema.Type<typeof TextVerbosity>
|
||||
readonly serviceTier?: ServiceTier
|
||||
}
|
||||
|
||||
export const resolve = (request: LLMRequest): Resolved => {
|
||||
const input = request.providerOptions?.[request.model.route.providerMetadataKey ?? "openresponses"]
|
||||
const include = Array.isArray(input?.include)
|
||||
? input.include.filter((entry): entry is ResponseIncludable => INCLUDABLES.has(entry))
|
||||
: []
|
||||
const reasoningSummary = input?.reasoningSummary
|
||||
return {
|
||||
instructions: typeof input?.instructions === "string" ? input.instructions : undefined,
|
||||
store: typeof input?.store === "boolean" ? input.store : undefined,
|
||||
reasoningEffort: typeof input?.reasoningEffort === "string" ? input.reasoningEffort : undefined,
|
||||
reasoningSummary:
|
||||
reasoningSummary === "auto" || reasoningSummary === "concise" || reasoningSummary === "detailed"
|
||||
? reasoningSummary
|
||||
: undefined,
|
||||
include: include.length > 0 ? include : undefined,
|
||||
textVerbosity: isTextVerbosity(input?.textVerbosity) ? input.textVerbosity : undefined,
|
||||
serviceTier: isServiceTier(input?.serviceTier) ? input.serviceTier : undefined,
|
||||
}
|
||||
}
|
||||
|
||||
export * as OpenResponsesOptions from "./open-responses-options.js"
|
||||
@@ -1,20 +0,0 @@
|
||||
import { Schema } from "effect"
|
||||
|
||||
const dimensions = (value: string) => {
|
||||
const match = /^(\d+)x(\d+)$/.exec(value)
|
||||
if (!match) return undefined
|
||||
return { width: Number(match[1]), height: Number(match[2]) }
|
||||
}
|
||||
|
||||
export const Size = Schema.String.check(
|
||||
Schema.makeFilter((value) => {
|
||||
if (value === "auto") return undefined
|
||||
const parsed = dimensions(value)
|
||||
if (!parsed) return "image size must be `auto` or `{width}x{height}`"
|
||||
return parsed.width > 0 && parsed.height > 0 ? undefined : "image dimensions must be positive integers"
|
||||
}),
|
||||
)
|
||||
|
||||
export const OpenAIImage = {
|
||||
Size,
|
||||
} as const
|
||||
@@ -1,23 +0,0 @@
|
||||
import { ReasoningEfforts } from "../../schema/index.js"
|
||||
import { OpenResponsesOptions } from "./open-responses-options.js"
|
||||
|
||||
export const OpenAIReasoningEfforts = ReasoningEfforts
|
||||
export type OpenAIReasoningEffort = string
|
||||
|
||||
// Mirrors OpenAI's `ResponseIncludable` union from the official SDK. Keep this
|
||||
// in lockstep with `openai-node/src/resources/responses/responses.ts`.
|
||||
export const OpenAIResponseIncludables = OpenResponsesOptions.ResponseIncludables
|
||||
export type OpenAIResponseIncludable = OpenResponsesOptions.ResponseIncludable
|
||||
export const OpenAIServiceTiers = OpenResponsesOptions.ServiceTiers
|
||||
export type OpenAIServiceTier = OpenResponsesOptions.ServiceTier
|
||||
|
||||
export const OpenAIReasoningEffort = OpenResponsesOptions.ReasoningEffort
|
||||
export const OpenAITextVerbosity = OpenResponsesOptions.TextVerbositySchema
|
||||
export const OpenAIResponseIncludable = OpenResponsesOptions.ResponseIncludableSchema
|
||||
export const OpenAIServiceTier = OpenResponsesOptions.ServiceTierSchema
|
||||
|
||||
export const isReasoningEffort = (effort: unknown): effort is OpenAIReasoningEffort => typeof effort === "string"
|
||||
|
||||
export const resolve = OpenResponsesOptions.resolve
|
||||
|
||||
export * as OpenAIOptions from "./openai-options.js"
|
||||
@@ -1,202 +0,0 @@
|
||||
import { Effect, Encoding, Schema } from "effect"
|
||||
import { Headers, HttpClientRequest } from "effect/unstable/http"
|
||||
import { GeneratedImage, ImageModel, ImageResponse, type ImageRequestFor, type ImageRoute } from "../image.js"
|
||||
import { Auth, type Definition as AuthDefinition } from "../route/auth.js"
|
||||
import {
|
||||
InvalidProviderOutputReason,
|
||||
AIError,
|
||||
Usage,
|
||||
mergeHttpOptions,
|
||||
mergeJsonRecords,
|
||||
type HttpOptions,
|
||||
} from "../schema/index.js"
|
||||
import { ProviderShared, optionalNull } from "./shared.js"
|
||||
import { ImageInputs } from "./utils/image-input.js"
|
||||
|
||||
const ADAPTER = "xai-images"
|
||||
export const DEFAULT_BASE_URL = "https://api.x.ai/v1"
|
||||
export const PATH = "/images/generations"
|
||||
export const EDIT_PATH = "/images/edits"
|
||||
|
||||
export type XAIImageString<Known extends string> = Known | (string & {})
|
||||
|
||||
export type XAIImageOptions = {
|
||||
readonly n?: number
|
||||
readonly aspectRatio?: XAIImageString<
|
||||
| "1:1"
|
||||
| "3:4"
|
||||
| "4:3"
|
||||
| "9:16"
|
||||
| "16:9"
|
||||
| "2:3"
|
||||
| "3:2"
|
||||
| "9:19.5"
|
||||
| "19.5:9"
|
||||
| "9:20"
|
||||
| "20:9"
|
||||
| "1:2"
|
||||
| "2:1"
|
||||
| "auto"
|
||||
>
|
||||
readonly aspect_ratio?: XAIImageString<
|
||||
| "1:1"
|
||||
| "3:4"
|
||||
| "4:3"
|
||||
| "9:16"
|
||||
| "16:9"
|
||||
| "2:3"
|
||||
| "3:2"
|
||||
| "9:19.5"
|
||||
| "19.5:9"
|
||||
| "9:20"
|
||||
| "20:9"
|
||||
| "1:2"
|
||||
| "2:1"
|
||||
| "auto"
|
||||
>
|
||||
readonly resolution?: XAIImageString<"1k" | "2k">
|
||||
readonly responseFormat?: XAIImageString<"url" | "b64_json">
|
||||
readonly response_format?: XAIImageString<"url" | "b64_json">
|
||||
} & Record<string, unknown>
|
||||
|
||||
type XAIImageBody = Record<string, unknown> & {
|
||||
readonly model: string
|
||||
readonly prompt: string
|
||||
}
|
||||
|
||||
const XAIImageResponse = Schema.Struct({
|
||||
data: Schema.Array(
|
||||
Schema.Struct({
|
||||
b64_json: optionalNull(Schema.String),
|
||||
url: optionalNull(Schema.String),
|
||||
revised_prompt: optionalNull(Schema.String),
|
||||
mime_type: optionalNull(Schema.String),
|
||||
}),
|
||||
),
|
||||
usage: Schema.optional(Schema.Unknown),
|
||||
})
|
||||
|
||||
export interface ModelInput {
|
||||
readonly id: string
|
||||
readonly auth: AuthDefinition
|
||||
readonly baseURL?: string
|
||||
readonly headers?: Record<string, string>
|
||||
readonly http?: HttpOptions
|
||||
}
|
||||
|
||||
const nativeOptions = (options: XAIImageOptions | undefined) => {
|
||||
if (!options) return undefined
|
||||
const { aspectRatio, responseFormat, ...native } = options
|
||||
return {
|
||||
aspect_ratio: aspectRatio,
|
||||
response_format: responseFormat,
|
||||
...native,
|
||||
}
|
||||
}
|
||||
|
||||
const invalidOutput = (message: string) =>
|
||||
new AIError({
|
||||
module: ADAPTER,
|
||||
method: "generate",
|
||||
reason: new InvalidProviderOutputReason({ message, route: ADAPTER }),
|
||||
})
|
||||
|
||||
const applyQuery = (url: string, query: Record<string, string> | undefined) => {
|
||||
if (!query) return url
|
||||
const next = new URL(url)
|
||||
Object.entries(query).forEach(([key, value]) => next.searchParams.set(key, value))
|
||||
return next.toString()
|
||||
}
|
||||
|
||||
export const model = (input: ModelInput) => {
|
||||
const route: ImageRoute<XAIImageOptions> = {
|
||||
id: ADAPTER,
|
||||
generate: Effect.fn("XAIImages.generate")(function* (request: ImageRequestFor<XAIImageOptions>, execute) {
|
||||
const http = mergeHttpOptions(request.model.http, request.http)
|
||||
const imageReferences = (request.images ?? []).map((image) => {
|
||||
if (image.type === "bytes") return { url: ImageInputs.dataUrl(image), type: "image_url" as const }
|
||||
if (image.type === "url") return { url: image.url, type: "image_url" as const }
|
||||
if (image.type === "file-id") return { file_id: image.id }
|
||||
return undefined
|
||||
})
|
||||
if (imageReferences.some((image) => image === undefined))
|
||||
return yield* ImageInputs.invalid(ADAPTER, "xAI Images accepts image URLs, data URLs, bytes, and file IDs")
|
||||
const requestBody = mergeJsonRecords(
|
||||
{
|
||||
model: request.model.id,
|
||||
prompt: request.prompt,
|
||||
image: imageReferences.length === 1 ? imageReferences[0] : undefined,
|
||||
images: imageReferences.length > 1 ? imageReferences : undefined,
|
||||
},
|
||||
nativeOptions(request.options),
|
||||
http?.body,
|
||||
) as XAIImageBody
|
||||
const text = ProviderShared.encodeJson(requestBody)
|
||||
const url = applyQuery(
|
||||
`${(input.baseURL ?? DEFAULT_BASE_URL).replace(/\/$/, "")}${imageReferences.length === 0 ? PATH : EDIT_PATH}`,
|
||||
http?.query,
|
||||
)
|
||||
const headers = yield* Auth.toEffect(input.auth)({
|
||||
request,
|
||||
method: "POST",
|
||||
url,
|
||||
body: text,
|
||||
headers: Headers.fromInput({ ...input.headers, ...http?.headers }),
|
||||
})
|
||||
const response = yield* execute(
|
||||
HttpClientRequest.post(url).pipe(
|
||||
HttpClientRequest.setHeaders(headers),
|
||||
HttpClientRequest.bodyText(text, "application/json"),
|
||||
),
|
||||
)
|
||||
const payload = yield* response.json.pipe(
|
||||
Effect.mapError(() => invalidOutput("Failed to read the xAI Images response")),
|
||||
)
|
||||
const decoded = yield* Schema.decodeUnknownEffect(XAIImageResponse)(payload).pipe(
|
||||
Effect.mapError(() => invalidOutput("xAI Images returned an invalid response")),
|
||||
)
|
||||
const images = yield* Effect.forEach(decoded.data, (item, index) => {
|
||||
const mediaType = item.mime_type ?? "application/octet-stream"
|
||||
if (item.b64_json)
|
||||
return Effect.fromResult(Encoding.decodeBase64(item.b64_json)).pipe(
|
||||
Effect.mapError(() => invalidOutput(`xAI Images result ${index} contains invalid base64 data`)),
|
||||
Effect.map(
|
||||
(data) =>
|
||||
new GeneratedImage({
|
||||
mediaType,
|
||||
data,
|
||||
providerMetadata:
|
||||
item.revised_prompt === undefined || item.revised_prompt === null
|
||||
? undefined
|
||||
: { xai: { revisedPrompt: item.revised_prompt } },
|
||||
}),
|
||||
),
|
||||
)
|
||||
if (item.url)
|
||||
return Effect.succeed(
|
||||
new GeneratedImage({
|
||||
mediaType,
|
||||
data: item.url,
|
||||
providerMetadata:
|
||||
item.revised_prompt === undefined || item.revised_prompt === null
|
||||
? undefined
|
||||
: { xai: { revisedPrompt: item.revised_prompt } },
|
||||
}),
|
||||
)
|
||||
return Effect.fail(invalidOutput(`xAI Images result ${index} has neither image data nor a URL`))
|
||||
})
|
||||
if (images.length === 0) return yield* invalidOutput("xAI Images returned no images")
|
||||
const usage = ProviderShared.isRecord(decoded.usage) ? decoded.usage : undefined
|
||||
return new ImageResponse({
|
||||
images,
|
||||
usage: usage === undefined ? undefined : new Usage({ providerMetadata: { xai: usage } }),
|
||||
providerMetadata: usage === undefined ? undefined : { xai: { usage } },
|
||||
})
|
||||
}),
|
||||
}
|
||||
return ImageModel.make<XAIImageOptions>({ id: input.id, provider: "xai", route, http: input.http })
|
||||
}
|
||||
|
||||
export const XAIImages = {
|
||||
model,
|
||||
} as const
|
||||
@@ -1,138 +0,0 @@
|
||||
import { Effect, Schema } from "effect"
|
||||
import { Headers, HttpClientRequest } from "effect/unstable/http"
|
||||
import { GeneratedImage, ImageModel, ImageResponse, type ImageRequestFor, type ImageRoute } from "../image.js"
|
||||
import { Auth, type Definition as AuthDefinition } from "../route/auth.js"
|
||||
import {
|
||||
InvalidProviderOutputReason,
|
||||
AIError,
|
||||
mergeHttpOptions,
|
||||
mergeJsonRecords,
|
||||
type HttpOptions,
|
||||
} from "../schema/index.js"
|
||||
import { ProviderShared } from "./shared.js"
|
||||
import { ImageInputs } from "./utils/image-input.js"
|
||||
|
||||
const ADAPTER = "zai-images"
|
||||
export const DEFAULT_BASE_URL = "https://api.z.ai/api/paas/v4"
|
||||
export const PATH = "/images/generations"
|
||||
|
||||
export type ZAIImageString<Known extends string> = Known | (string & {})
|
||||
|
||||
export type ZAIImageOptions = {
|
||||
readonly size?: ZAIImageString<
|
||||
"1024x1024" | "768x1344" | "864x1152" | "1344x768" | "1152x864" | "1440x720" | "720x1440"
|
||||
>
|
||||
readonly quality?: ZAIImageString<"hd" | "standard">
|
||||
readonly userID?: string
|
||||
} & Record<string, unknown>
|
||||
|
||||
type ZAIImageBody = Record<string, unknown> & {
|
||||
readonly model: string
|
||||
readonly prompt: string
|
||||
}
|
||||
|
||||
const ZAIImageResponse = Schema.Struct({
|
||||
created: Schema.optional(Schema.Int),
|
||||
id: Schema.optional(Schema.String),
|
||||
request_id: Schema.optional(Schema.String),
|
||||
data: Schema.Array(Schema.Struct({ url: Schema.String })),
|
||||
content_filter: Schema.optional(
|
||||
Schema.Array(
|
||||
Schema.Struct({
|
||||
role: Schema.optional(Schema.String),
|
||||
level: Schema.optional(Schema.Number),
|
||||
}),
|
||||
),
|
||||
),
|
||||
})
|
||||
|
||||
export interface ModelInput {
|
||||
readonly id: string
|
||||
readonly auth: AuthDefinition
|
||||
readonly baseURL?: string
|
||||
readonly headers?: Record<string, string>
|
||||
readonly http?: HttpOptions
|
||||
}
|
||||
|
||||
const nativeOptions = (options: ZAIImageOptions | undefined) => {
|
||||
if (!options) return undefined
|
||||
const { userID, ...native } = options
|
||||
return {
|
||||
user_id: userID,
|
||||
...native,
|
||||
}
|
||||
}
|
||||
|
||||
const invalidOutput = (message: string) =>
|
||||
new AIError({
|
||||
module: ADAPTER,
|
||||
method: "generate",
|
||||
reason: new InvalidProviderOutputReason({ message, route: ADAPTER }),
|
||||
})
|
||||
|
||||
const applyQuery = (url: string, query: Record<string, string> | undefined) => {
|
||||
if (!query) return url
|
||||
const next = new URL(url)
|
||||
Object.entries(query).forEach(([key, value]) => next.searchParams.set(key, value))
|
||||
return next.toString()
|
||||
}
|
||||
|
||||
export const model = (input: ModelInput) => {
|
||||
const route: ImageRoute<ZAIImageOptions> = {
|
||||
id: ADAPTER,
|
||||
generate: Effect.fn("ZAIImages.generate")(function* (request: ImageRequestFor<ZAIImageOptions>, execute) {
|
||||
if ((request.images?.length ?? 0) > 0)
|
||||
return yield* ImageInputs.invalid(ADAPTER, "Z.ai hosted image generation does not support image inputs")
|
||||
const http = mergeHttpOptions(request.model.http, request.http)
|
||||
const requestBody = mergeJsonRecords(
|
||||
{ model: request.model.id, prompt: request.prompt },
|
||||
nativeOptions(request.options),
|
||||
http?.body,
|
||||
) as ZAIImageBody
|
||||
const text = ProviderShared.encodeJson(requestBody)
|
||||
const url = applyQuery(`${(input.baseURL ?? DEFAULT_BASE_URL).replace(/\/$/, "")}${PATH}`, http?.query)
|
||||
const headers = yield* Auth.toEffect(input.auth)({
|
||||
request,
|
||||
method: "POST",
|
||||
url,
|
||||
body: text,
|
||||
headers: Headers.fromInput({ ...input.headers, ...http?.headers }),
|
||||
})
|
||||
const response = yield* execute(
|
||||
HttpClientRequest.post(url).pipe(
|
||||
HttpClientRequest.setHeaders(headers),
|
||||
HttpClientRequest.bodyText(text, "application/json"),
|
||||
),
|
||||
)
|
||||
const payload = yield* response.json.pipe(
|
||||
Effect.mapError(() => invalidOutput("Failed to read the Z.ai Images response")),
|
||||
)
|
||||
const decoded = yield* Schema.decodeUnknownEffect(ZAIImageResponse)(payload).pipe(
|
||||
Effect.mapError(() => invalidOutput("Z.ai Images returned an invalid response")),
|
||||
)
|
||||
if (decoded.data.length === 0) return yield* invalidOutput("Z.ai Images returned no images")
|
||||
return new ImageResponse({
|
||||
images: decoded.data.map(
|
||||
(item) =>
|
||||
new GeneratedImage({
|
||||
mediaType: "application/octet-stream",
|
||||
data: item.url,
|
||||
}),
|
||||
),
|
||||
providerMetadata: {
|
||||
zai: {
|
||||
created: decoded.created,
|
||||
id: decoded.id,
|
||||
requestID: decoded.request_id,
|
||||
contentFilter: decoded.content_filter,
|
||||
},
|
||||
},
|
||||
})
|
||||
}),
|
||||
}
|
||||
return ImageModel.make<ZAIImageOptions>({ id: input.id, provider: "zai", route, http: input.http })
|
||||
}
|
||||
|
||||
export const ZAIImages = {
|
||||
model,
|
||||
} as const
|
||||
@@ -1,164 +0,0 @@
|
||||
import { Option, Schema } from "effect"
|
||||
import {
|
||||
AuthenticationReason,
|
||||
ContentPolicyReason,
|
||||
InvalidRequestReason,
|
||||
AIError,
|
||||
ProviderErrorEvent,
|
||||
ProviderInternalReason,
|
||||
QuotaExceededReason,
|
||||
RateLimitReason,
|
||||
UnknownProviderReason,
|
||||
type HttpContext,
|
||||
type HttpRateLimitDetails,
|
||||
type ProviderMetadata,
|
||||
} from "./schema/index.js"
|
||||
|
||||
const patterns = [
|
||||
/prompt is too long/i,
|
||||
/input is too long for requested model/i,
|
||||
/exceeds the context window/i,
|
||||
/exceeds (?:the )?(?:model'?s )?maximum context length(?: of [\d,]+ tokens?|\s*\([\d,]+\))/i,
|
||||
/input token count.*exceeds the maximum/i,
|
||||
/tokens in request more than max tokens allowed/i,
|
||||
/maximum prompt length is \d+/i,
|
||||
/reduce the length of the messages/i,
|
||||
/maximum context length is \d+ tokens/i,
|
||||
/exceeds (?:the )?maximum allowed input length of [\d,]+ tokens?/i,
|
||||
/input \(\d+ tokens\) is longer than the model'?s context length \(\d+ tokens\)/i,
|
||||
/exceeds the limit of \d+/i,
|
||||
/exceeds the available context size/i,
|
||||
/greater than the context length/i,
|
||||
/context window exceeds limit/i,
|
||||
/exceeded model token limit/i,
|
||||
/context[_ ]length[_ ]exceeded/i,
|
||||
/context length is only \d+ tokens/i,
|
||||
/input length.*exceeds.*context length/i,
|
||||
/prompt too long; exceeded (?:max )?context length/i,
|
||||
/too large for model with \d+ maximum context length/i,
|
||||
/prompt has [\d,]+ tokens?, but the configured context size is [\d,]+ tokens?/i,
|
||||
/model_context_window_exceeded/i,
|
||||
/too many tokens/i,
|
||||
/token limit exceeded/i,
|
||||
]
|
||||
|
||||
const payloadPatterns = [/request_too_large/i, /request entity too large/i, /payload too large/i, /request too large/i]
|
||||
|
||||
const exclusions = [/^(throttling error|service unavailable):/i, /rate limit/i, /too many requests/i]
|
||||
|
||||
export const isContextOverflow = (message: string) =>
|
||||
!exclusions.some((pattern) => pattern.test(message)) &&
|
||||
(patterns.some((pattern) => pattern.test(message)) || /^400\s*(status code)?\s*\(no body\)/i.test(message))
|
||||
|
||||
export const isPayloadTooLarge = (message: string) => payloadPatterns.some((pattern) => pattern.test(message))
|
||||
|
||||
export const isContextOverflowFailure = (failure: unknown) =>
|
||||
failure instanceof AIError
|
||||
? failure.reason._tag === "InvalidRequest" && failure.reason.classification === "context-overflow"
|
||||
: Schema.is(ProviderErrorEvent)(failure) && failure.classification === "context-overflow"
|
||||
|
||||
const decodeJson = Schema.decodeUnknownOption(Schema.UnknownFromJsonString)
|
||||
const QUOTA_CODES = new Set(["insufficient_quota", "usage_not_included", "billing_error"])
|
||||
const SERVER_CODES = new Set([
|
||||
"api_error",
|
||||
"internal_error",
|
||||
"internalserverexception",
|
||||
"modelstreamerrorexception",
|
||||
"overloaded_error",
|
||||
"server_error",
|
||||
"server_is_overloaded",
|
||||
"slow_down",
|
||||
"serviceunavailableexception",
|
||||
])
|
||||
const INVALID_REQUEST_CODES = new Set(["invalid_prompt", "invalid_request_error", "validationexception"])
|
||||
const RATE_LIMIT_TEXT = /rate increased too quickly|rate[-_\s]?limit|too[_\s]?many[_\s]?requests/i
|
||||
const QUOTA_TEXT = /insufficient[-_\s]?quota|quota[-_\s]?exceeded/i
|
||||
const CONTENT_POLICY_TEXT = /content[-_\s]?policy|content_filter|safety/i
|
||||
|
||||
export interface ProviderFailure {
|
||||
readonly message: string
|
||||
readonly status?: number | undefined
|
||||
readonly code?: string | undefined
|
||||
readonly retryAfterMs?: number | undefined
|
||||
readonly rateLimit?: HttpRateLimitDetails | undefined
|
||||
readonly http?: HttpContext | undefined
|
||||
readonly providerMetadata?: ProviderMetadata | undefined
|
||||
}
|
||||
|
||||
// Keep HTTP failures and provider-reported stream failures on one typed path so
|
||||
// session retry policy never needs provider-specific string matching.
|
||||
export function classifyProviderFailure(input: ProviderFailure): AIError["reason"] {
|
||||
const body = input.http?.body ?? ""
|
||||
const codes = [input.code, ...providerCodes(body), ...providerCodes(input.message)]
|
||||
.filter((code): code is string => code !== undefined)
|
||||
.map((code) => code.toLowerCase())
|
||||
const text = body || input.message
|
||||
const common = { message: input.message, providerMetadata: input.providerMetadata, http: input.http }
|
||||
const clientScoped = input.status === undefined || (input.status >= 400 && input.status < 500)
|
||||
|
||||
if (
|
||||
clientScoped &&
|
||||
(codes.includes("context_length_exceeded") ||
|
||||
codes.includes("model_context_window_exceeded") ||
|
||||
isContextOverflow(text))
|
||||
)
|
||||
return new InvalidRequestReason({ ...common, classification: "context-overflow" })
|
||||
if (input.status === 413 || isPayloadTooLarge(text))
|
||||
return new InvalidRequestReason({ ...common, classification: "payload-too-large" })
|
||||
if (CONTENT_POLICY_TEXT.test(text)) return new ContentPolicyReason(common)
|
||||
if (codes.some((code) => QUOTA_CODES.has(code)) || (input.status === 429 && QUOTA_TEXT.test(text)))
|
||||
return new QuotaExceededReason(common)
|
||||
if (input.status === 401) return new AuthenticationReason({ ...common, kind: "invalid" })
|
||||
if (input.status === 403) return new AuthenticationReason({ ...common, kind: "insufficient-permissions" })
|
||||
if (codes.includes("authentication_error")) return new AuthenticationReason({ ...common, kind: "invalid" })
|
||||
if (codes.includes("permission_error"))
|
||||
return new AuthenticationReason({ ...common, kind: "insufficient-permissions" })
|
||||
if (
|
||||
codes.some((code) => code.includes("rate_limit") || code === "too_many_requests" || code === "throttlingexception")
|
||||
)
|
||||
return new RateLimitReason({
|
||||
...common,
|
||||
retryAfterMs: input.retryAfterMs,
|
||||
rateLimit: input.rateLimit,
|
||||
})
|
||||
if (RATE_LIMIT_TEXT.test(text))
|
||||
return new RateLimitReason({
|
||||
...common,
|
||||
retryAfterMs: input.retryAfterMs,
|
||||
rateLimit: input.rateLimit,
|
||||
})
|
||||
if (codes.some((code) => SERVER_CODES.has(code) || code.includes("exhausted") || code.includes("unavailable")))
|
||||
return new ProviderInternalReason({
|
||||
...common,
|
||||
status: input.status,
|
||||
retryAfterMs: input.retryAfterMs,
|
||||
})
|
||||
if (input.status === 429) {
|
||||
return new RateLimitReason({
|
||||
...common,
|
||||
retryAfterMs: input.retryAfterMs,
|
||||
rateLimit: input.rateLimit,
|
||||
})
|
||||
}
|
||||
if (input.status === 408 || input.status === 409 || (input.status !== undefined && input.status >= 500))
|
||||
return new ProviderInternalReason({
|
||||
...common,
|
||||
status: input.status,
|
||||
retryAfterMs: input.retryAfterMs,
|
||||
})
|
||||
if (codes.some((code) => INVALID_REQUEST_CODES.has(code))) return new InvalidRequestReason(common)
|
||||
if (input.status === 400 || input.status === 404 || input.status === 413 || input.status === 422)
|
||||
return new InvalidRequestReason(common)
|
||||
return new UnknownProviderReason({ ...common, status: input.status })
|
||||
}
|
||||
|
||||
function providerCodes(value: string) {
|
||||
const decoded = Option.getOrUndefined(decodeJson(value))
|
||||
if (!isRecord(decoded)) return []
|
||||
const error = isRecord(decoded.error) ? decoded.error : undefined
|
||||
return [decoded.code, error?.code, error?.type].filter((value): value is string => typeof value === "string")
|
||||
}
|
||||
|
||||
function isRecord(value: unknown): value is Record<string, unknown> {
|
||||
return typeof value === "object" && value !== null
|
||||
}
|
||||
@@ -1,21 +0,0 @@
|
||||
import type { LanguageModel, ProviderOptions } from "./schema/index.js"
|
||||
|
||||
export interface Settings extends Readonly<Record<string, unknown>> {
|
||||
readonly baseURL?: string
|
||||
readonly headers?: Readonly<Record<string, string>>
|
||||
readonly body?: Readonly<Record<string, unknown>>
|
||||
readonly limits?: {
|
||||
readonly context: number
|
||||
readonly input?: number
|
||||
readonly output: number
|
||||
}
|
||||
}
|
||||
|
||||
export interface Definition<
|
||||
ProviderSettings extends Settings = Settings,
|
||||
Options extends ProviderOptions = ProviderOptions,
|
||||
> {
|
||||
readonly model: (modelID: string, settings: ProviderSettings) => LanguageModel<Options>
|
||||
}
|
||||
|
||||
export * as ProviderPackage from "./provider-package.js"
|
||||
@@ -1,36 +0,0 @@
|
||||
import type { LanguageModel, ModelID, ProviderID } from "./schema/index.js"
|
||||
|
||||
export type LanguageModelOptions = Pick<LanguageModel.Input, "defaults" | "compatibility">
|
||||
|
||||
/**
|
||||
* Advanced structural provider definition helper. Built-in providers should
|
||||
* prefer explicit `configure(options).model(id)` facades so deployment config is
|
||||
* chosen before model selection. The optional `apis` map remains for external
|
||||
* structural providers that expose multiple route selectors behind one provider.
|
||||
*/
|
||||
export type LanguageModelFactory<Options extends LanguageModelOptions = LanguageModelOptions> = (
|
||||
id: string | ModelID,
|
||||
options?: Options,
|
||||
) => LanguageModel
|
||||
|
||||
type AnyLanguageModelFactory = (...args: never[]) => LanguageModel
|
||||
|
||||
export interface Definition<Factory extends AnyLanguageModelFactory = LanguageModelFactory> {
|
||||
readonly id: ProviderID
|
||||
readonly model: Factory
|
||||
readonly apis?: Record<string, AnyLanguageModelFactory>
|
||||
}
|
||||
|
||||
type DefinitionShape = {
|
||||
readonly id: ProviderID
|
||||
readonly model: (...args: never[]) => LanguageModel
|
||||
readonly apis?: Record<string, (...args: never[]) => LanguageModel>
|
||||
}
|
||||
|
||||
type NoExtraFields<Input, Shape> = Input & Record<Exclude<keyof Input, keyof Shape>, never>
|
||||
|
||||
export const make = <DefinitionType extends DefinitionShape>(
|
||||
definition: NoExtraFields<DefinitionType, DefinitionShape>,
|
||||
) => definition
|
||||
|
||||
export * as Provider from "./provider.js"
|
||||
@@ -1 +0,0 @@
|
||||
export * from "./providers/index.js"
|
||||
@@ -1,107 +0,0 @@
|
||||
import { Auth } from "../route/auth.js"
|
||||
import type { Route as RouteDef, RouteDefaultsInput } from "../route/client.js"
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { OpenAIChat } from "../protocols/openai-chat.js"
|
||||
import { OpenAIResponses } from "../protocols/openai-responses.js"
|
||||
import { BedrockAuth, type Credentials } from "../protocols/utils/bedrock-auth.js"
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
import { withOpenAIOptions, type OpenAIProviderOptionsInput } from "./openai-options.js"
|
||||
|
||||
export const id = ProviderID.make("amazon-bedrock")
|
||||
|
||||
export type Config = RouteDefaultsInput & {
|
||||
readonly apiKey?: string
|
||||
readonly baseURL?: string
|
||||
readonly credentials?: Credentials
|
||||
readonly region?: string
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
export interface Settings extends ProviderPackage.Settings {
|
||||
readonly apiKey?: string
|
||||
readonly auth?: "bearer" | "sigv4"
|
||||
readonly baseURL?: string
|
||||
readonly credentials?: Credentials
|
||||
readonly region?: string
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
const responsesRoute = OpenAIResponses.route.with({
|
||||
id: "bedrock-mantle-responses",
|
||||
provider: id,
|
||||
})
|
||||
|
||||
const chatRoute = OpenAIChat.route.with({
|
||||
id: "bedrock-mantle-chat",
|
||||
provider: id,
|
||||
})
|
||||
|
||||
export const routes = [responsesRoute, chatRoute]
|
||||
|
||||
const configuredRoute = <Body, Prepared>(route: RouteDef<Body, Prepared>, input: Config) => {
|
||||
const region = input.region ?? input.credentials?.region ?? "us-east-1"
|
||||
const credentials = input.credentials === undefined ? undefined : { ...input.credentials, region }
|
||||
return route.with({
|
||||
endpoint: { baseURL: input.baseURL ?? `https://bedrock-mantle.${region}.api.aws/v1` },
|
||||
auth:
|
||||
input.apiKey === undefined
|
||||
? BedrockAuth.sigV4(credentials, { service: "bedrock-mantle", name: "Bedrock Mantle" })
|
||||
: Auth.bearer(input.apiKey),
|
||||
})
|
||||
}
|
||||
|
||||
const defaults = (input: Config) => {
|
||||
const { apiKey: _, baseURL: _baseURL, credentials: _credentials, region: _region, ...rest } = input
|
||||
return rest
|
||||
}
|
||||
|
||||
export const configure = (input: Config = {}) => {
|
||||
const configuredResponsesRoute = configuredRoute(responsesRoute, input)
|
||||
const configuredChatRoute = configuredRoute(chatRoute, input)
|
||||
const modelDefaults = defaults(input)
|
||||
const responses = (modelID: string | ModelID) =>
|
||||
configuredResponsesRoute
|
||||
.with(withOpenAIOptions(modelID, modelDefaults))
|
||||
.model<OpenAIProviderOptionsInput>({ id: modelID })
|
||||
const chat = (modelID: string | ModelID) =>
|
||||
configuredChatRoute
|
||||
.with(withOpenAIOptions(modelID, modelDefaults))
|
||||
.model<OpenAIProviderOptionsInput>({ id: modelID })
|
||||
|
||||
return {
|
||||
id,
|
||||
model: chat,
|
||||
chat,
|
||||
responses,
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = configure()
|
||||
|
||||
const config = (settings: Settings): Config => {
|
||||
if (settings.auth === "bearer" && settings.apiKey === undefined)
|
||||
throw new Error("Amazon Bedrock Mantle bearer auth requires apiKey")
|
||||
if (settings.auth === "sigv4" && settings.apiKey !== undefined)
|
||||
throw new Error("Amazon Bedrock Mantle SigV4 auth does not accept apiKey")
|
||||
return {
|
||||
apiKey: settings.auth === "sigv4" ? undefined : settings.apiKey,
|
||||
baseURL: settings.baseURL,
|
||||
credentials: settings.credentials,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
providerOptions: settings.providerOptions,
|
||||
region: settings.region,
|
||||
}
|
||||
}
|
||||
|
||||
export const chatModel: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) => configure(config(settings)).chat(modelID)
|
||||
export const responsesModel: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) => configure(config(settings)).responses(modelID)
|
||||
export const model = chatModel
|
||||
@@ -1,2 +0,0 @@
|
||||
export { chatModel as model } from "../amazon-bedrock-mantle.js"
|
||||
export type { Settings } from "../amazon-bedrock-mantle.js"
|
||||
@@ -1,2 +0,0 @@
|
||||
export { chatModel as model } from "../../amazon-bedrock-mantle.js"
|
||||
export type { Settings } from "../../amazon-bedrock-mantle.js"
|
||||
@@ -1,2 +0,0 @@
|
||||
export { responsesModel as model } from "../../amazon-bedrock-mantle.js"
|
||||
export type { Settings } from "../../amazon-bedrock-mantle.js"
|
||||
@@ -1,77 +0,0 @@
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { AnthropicMessages } from "../protocols/anthropic-messages.js"
|
||||
import { Auth } from "../route/auth.js"
|
||||
import type { ProviderAuthOption } from "../route/auth-options.js"
|
||||
import type { RouteDefaultsInput } from "../route/client.js"
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
|
||||
export type AnthropicOptionsInput = AnthropicMessages.OptionsInput
|
||||
export type AnthropicProviderOptionsInput = AnthropicMessages.ProviderOptionsInput
|
||||
export type AnthropicThinkingInput = AnthropicMessages.ThinkingInput
|
||||
|
||||
export const id = ProviderID.make("anthropic-compatible")
|
||||
|
||||
export type Config = RouteDefaultsInput &
|
||||
ProviderAuthOption<"optional"> & {
|
||||
readonly provider?: string
|
||||
readonly baseURL: string
|
||||
readonly providerOptions?: AnthropicMessages.ProviderOptionsInput
|
||||
}
|
||||
|
||||
export type Settings = ProviderPackage.Settings &
|
||||
(
|
||||
| { readonly apiKey?: string; readonly authToken?: never }
|
||||
| { readonly apiKey?: never; readonly authToken?: string }
|
||||
) & {
|
||||
readonly baseURL: string
|
||||
readonly provider?: string
|
||||
readonly providerOptions?: AnthropicMessages.ProviderOptionsInput
|
||||
}
|
||||
|
||||
export const routes = [AnthropicMessages.route]
|
||||
|
||||
const auth = (input: ProviderAuthOption<"optional">) => {
|
||||
if ("auth" in input && input.auth) return input.auth
|
||||
return Auth.optional("apiKey" in input ? input.apiKey : undefined, "apiKey").pipe(Auth.header("x-api-key"))
|
||||
}
|
||||
|
||||
export const configure = (input: Config) => {
|
||||
if (!input.baseURL) throw new Error("Anthropic-compatible providers require a baseURL")
|
||||
const provider = input.provider ?? "anthropic-compatible"
|
||||
const { provider: _, baseURL, apiKey: _apiKey, auth: _auth, ...rest } = input
|
||||
const route = AnthropicMessages.route.with({
|
||||
...rest,
|
||||
provider,
|
||||
endpoint: { baseURL },
|
||||
auth: auth(input),
|
||||
})
|
||||
return {
|
||||
id: ProviderID.make(provider),
|
||||
model: (modelID: string | ModelID) => route.model<AnthropicMessages.ProviderOptionsInput>({ id: modelID }),
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = {
|
||||
id,
|
||||
configure,
|
||||
}
|
||||
|
||||
export const model: ProviderPackage.Definition<Settings, AnthropicMessages.ProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) => {
|
||||
if (settings.apiKey !== undefined && settings.authToken !== undefined)
|
||||
throw new Error("Anthropic-compatible apiKey cannot be combined with authToken")
|
||||
return configure({
|
||||
...(settings.authToken === undefined ? { apiKey: settings.apiKey } : { auth: Auth.bearer(settings.authToken) }),
|
||||
baseURL: settings.baseURL,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
provider: settings.provider,
|
||||
providerOptions: settings.providerOptions,
|
||||
}).model(modelID)
|
||||
}
|
||||
|
||||
export * as AnthropicCompatible from "./anthropic-compatible.js"
|
||||
@@ -1,69 +0,0 @@
|
||||
import type { RouteDefaultsInput } from "../route/client.js"
|
||||
import { Auth } from "../route/auth.js"
|
||||
import type { ProviderAuthOption } from "../route/auth-options.js"
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
import { AnthropicMessages } from "../protocols/anthropic-messages.js"
|
||||
import { AnthropicCompatible } from "./anthropic-compatible.js"
|
||||
|
||||
export type AnthropicOptionsInput = AnthropicMessages.OptionsInput
|
||||
export type AnthropicProviderOptionsInput = AnthropicMessages.ProviderOptionsInput
|
||||
export type AnthropicThinkingInput = AnthropicMessages.ThinkingInput
|
||||
|
||||
export const id = ProviderID.make("anthropic")
|
||||
|
||||
export const routes = [AnthropicMessages.route]
|
||||
|
||||
export type Config = RouteDefaultsInput &
|
||||
ProviderAuthOption<"optional"> & {
|
||||
readonly baseURL?: string
|
||||
readonly providerOptions?: AnthropicMessages.ProviderOptionsInput
|
||||
}
|
||||
|
||||
export type Settings = ProviderPackage.Settings &
|
||||
(
|
||||
| { readonly apiKey?: string; readonly authToken?: never }
|
||||
| { readonly apiKey?: never; readonly authToken?: string }
|
||||
) & {
|
||||
readonly baseURL?: string
|
||||
readonly providerOptions?: AnthropicMessages.ProviderOptionsInput
|
||||
}
|
||||
|
||||
const auth = (options: ProviderAuthOption<"optional">) => {
|
||||
if ("auth" in options && options.auth) return options.auth
|
||||
return Auth.optional("apiKey" in options ? options.apiKey : undefined, "apiKey")
|
||||
.orElse(Auth.config("ANTHROPIC_API_KEY"))
|
||||
.pipe(Auth.header("x-api-key"))
|
||||
}
|
||||
|
||||
export const configure = (input: Config = {}) => {
|
||||
const { apiKey: _, auth: _auth, baseURL, ...rest } = input
|
||||
const compatible = AnthropicCompatible.configure({
|
||||
...rest,
|
||||
auth: auth(input),
|
||||
baseURL: baseURL ?? AnthropicMessages.DEFAULT_BASE_URL,
|
||||
provider: id,
|
||||
})
|
||||
return {
|
||||
id,
|
||||
model: (modelID: string | ModelID) => compatible.model(modelID),
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = configure()
|
||||
export const model: ProviderPackage.Definition<Settings, AnthropicMessages.ProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) => {
|
||||
if (settings.apiKey !== undefined && settings.authToken !== undefined)
|
||||
throw new Error("Anthropic apiKey cannot be combined with authToken")
|
||||
return configure({
|
||||
...(settings.authToken === undefined ? { apiKey: settings.apiKey } : { auth: Auth.bearer(settings.authToken) }),
|
||||
baseURL: settings.baseURL,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
providerOptions: settings.providerOptions,
|
||||
}).model(modelID)
|
||||
}
|
||||
@@ -1,147 +0,0 @@
|
||||
import { Auth } from "../route/auth.js"
|
||||
import { type AtLeastOne, type ProviderAuthOption } from "../route/auth-options.js"
|
||||
import type { Route as RouteDef, RouteDefaultsInput } from "../route/client.js"
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
import * as OpenAIChat from "../protocols/openai-chat.js"
|
||||
import * as OpenAIResponses from "../protocols/openai-responses.js"
|
||||
import { ProviderShared } from "../protocols/shared.js"
|
||||
import { withOpenAIOptions, type OpenAIProviderOptionsInput } from "./openai-options.js"
|
||||
|
||||
export const id = ProviderID.make("azure")
|
||||
const routeAuth = Auth.remove("authorization")
|
||||
|
||||
// Azure needs the customer's resource URL; supply either `resourceName`
|
||||
// (helper builds the URL) or `baseURL` directly.
|
||||
type AzureURL = AtLeastOne<{ readonly resourceName: string; readonly baseURL: string }>
|
||||
|
||||
export type LanguageModelOptions = AzureURL &
|
||||
RouteDefaultsInput &
|
||||
ProviderAuthOption<"optional"> & {
|
||||
readonly apiVersion?: string
|
||||
readonly queryParams?: Record<string, string>
|
||||
readonly useDeploymentBasedUrls?: boolean
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
export type Config = LanguageModelOptions
|
||||
|
||||
export type Settings = ProviderPackage.Settings &
|
||||
AzureURL & {
|
||||
readonly apiKey?: string
|
||||
readonly apiVersion?: string
|
||||
readonly queryParams?: Readonly<Record<string, string>>
|
||||
readonly useDeploymentBasedUrls?: boolean
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
const resourceBaseURL = (resourceName: string) => `https://${resourceName.trim()}.openai.azure.com/openai`
|
||||
|
||||
const responsesRoute = OpenAIResponses.route.with({
|
||||
id: "azure-openai-responses",
|
||||
provider: id,
|
||||
auth: routeAuth,
|
||||
})
|
||||
|
||||
const chatRoute = OpenAIChat.route.with({
|
||||
id: "azure-openai-chat",
|
||||
provider: id,
|
||||
auth: routeAuth,
|
||||
})
|
||||
|
||||
export const routes = [responsesRoute, chatRoute]
|
||||
|
||||
const defaults = (input: Config) => {
|
||||
const {
|
||||
apiKey: _,
|
||||
apiVersion: _apiVersion,
|
||||
resourceName: _resourceName,
|
||||
useDeploymentBasedUrls: _useDeploymentBasedUrls,
|
||||
baseURL: _baseURL,
|
||||
queryParams: _queryParams,
|
||||
...rest
|
||||
} = input
|
||||
if ("auth" in rest) {
|
||||
const { auth: _, ...withoutAuth } = rest
|
||||
return withoutAuth
|
||||
}
|
||||
return rest
|
||||
}
|
||||
|
||||
const auth = (input: Config) => {
|
||||
if ("auth" in input && input.auth) return input.auth
|
||||
return Auth.remove("authorization").andThen(
|
||||
Auth.optional("apiKey" in input ? input.apiKey : undefined, "apiKey")
|
||||
.orElse(Auth.config("AZURE_OPENAI_API_KEY"))
|
||||
.pipe(Auth.header("api-key")),
|
||||
)
|
||||
}
|
||||
|
||||
const configuredRoute = <Body, Prepared>(route: RouteDef<Body, Prepared>, input: Config, modelID: string | ModelID) =>
|
||||
route.with({
|
||||
auth: auth(input),
|
||||
endpoint: endpoint(input, modelID),
|
||||
})
|
||||
|
||||
function endpoint(input: Config, modelID: string | ModelID) {
|
||||
const baseURL = ProviderShared.trimBaseUrl(input.baseURL ?? resourceBaseURL(input.resourceName!))
|
||||
const query = { "api-version": input.apiVersion ?? "v1", ...input.queryParams }
|
||||
|
||||
if (input.useDeploymentBasedUrls) return { baseURL: `${baseURL}/deployments/${modelID}`, query }
|
||||
if (input.baseURL !== undefined && !new URL(input.baseURL).hostname.endsWith(".openai.azure.com")) {
|
||||
return { baseURL, query: input.queryParams }
|
||||
}
|
||||
return { baseURL: `${baseURL}/v1`, query }
|
||||
}
|
||||
|
||||
export const configure = (input: Config) => {
|
||||
const modelDefaults = defaults(input)
|
||||
|
||||
const responses = (modelID: string | ModelID) =>
|
||||
configuredRoute(responsesRoute, input, modelID)
|
||||
.with(withOpenAIOptions(modelID, modelDefaults))
|
||||
.model<OpenAIProviderOptionsInput>({ id: modelID })
|
||||
|
||||
const chat = (modelID: string | ModelID) =>
|
||||
configuredRoute(chatRoute, input, modelID)
|
||||
.with(withOpenAIOptions(modelID, modelDefaults))
|
||||
.model<OpenAIProviderOptionsInput>({ id: modelID })
|
||||
|
||||
return {
|
||||
id,
|
||||
model: responses,
|
||||
responses,
|
||||
chat,
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = {
|
||||
id,
|
||||
configure,
|
||||
}
|
||||
|
||||
const config = (settings: Settings): Config => {
|
||||
const common = {
|
||||
apiKey: settings.apiKey,
|
||||
apiVersion: settings.apiVersion,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
providerOptions: settings.providerOptions,
|
||||
queryParams: settings.queryParams === undefined ? undefined : { ...settings.queryParams },
|
||||
useDeploymentBasedUrls: settings.useDeploymentBasedUrls,
|
||||
}
|
||||
if (settings.baseURL !== undefined) return { ...common, baseURL: settings.baseURL }
|
||||
if (settings.resourceName !== undefined) return { ...common, resourceName: settings.resourceName }
|
||||
throw new Error("Azure requires resourceName or baseURL")
|
||||
}
|
||||
|
||||
export const responsesModel: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) => configure(config(settings)).responses(modelID)
|
||||
export const chatModel: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) => configure(config(settings)).chat(modelID)
|
||||
export const model = responsesModel
|
||||
@@ -1,2 +0,0 @@
|
||||
export { chatModel as model } from "../azure.js"
|
||||
export type { Settings } from "../azure.js"
|
||||
@@ -1,2 +0,0 @@
|
||||
export { responsesModel as model } from "../azure.js"
|
||||
export type { Settings } from "../azure.js"
|
||||
@@ -1,133 +0,0 @@
|
||||
import type { Config, Redacted } from "effect"
|
||||
import * as OpenAICompatibleChat from "../protocols/openai-compatible-chat.js"
|
||||
import { Auth } from "../route/auth.js"
|
||||
import { AuthOptions, type AtLeastOne, type ProviderAuthOption } from "../route/auth-options.js"
|
||||
import type { RouteDefaultsInput } from "../route/client.js"
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
import type { OpenAIProviderOptionsInput } from "./openai-options.js"
|
||||
|
||||
export const aiGatewayID = ProviderID.make("cloudflare-ai-gateway")
|
||||
export const workersAIID = ProviderID.make("cloudflare-workers-ai")
|
||||
export const aiGatewayAuthEnvVars = ["CLOUDFLARE_API_TOKEN", "CF_AIG_TOKEN"] as const
|
||||
export const workersAIAuthEnvVars = ["CLOUDFLARE_API_KEY", "CLOUDFLARE_WORKERS_AI_TOKEN"] as const
|
||||
|
||||
type CloudflareSecret = string | Redacted.Redacted | Config.Config<string | Redacted.Redacted>
|
||||
|
||||
type GatewayURL = AtLeastOne<{
|
||||
readonly accountId: string
|
||||
readonly baseURL: string
|
||||
}> & {
|
||||
readonly gatewayId?: string
|
||||
}
|
||||
|
||||
export type AIGatewayOptions = GatewayURL &
|
||||
Omit<RouteDefaultsInput, "providerOptions"> &
|
||||
ProviderAuthOption<"optional"> & {
|
||||
/** Cloudflare AI Gateway authentication token. Sent as `cf-aig-authorization`. */
|
||||
readonly gatewayApiKey?: CloudflareSecret
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
type WorkersAIURL = AtLeastOne<{
|
||||
readonly accountId: string
|
||||
readonly baseURL: string
|
||||
}>
|
||||
|
||||
export type WorkersAIOptions = WorkersAIURL &
|
||||
Omit<RouteDefaultsInput, "providerOptions"> &
|
||||
ProviderAuthOption<"optional"> & {
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
export const aiGatewayBaseURL = (input: GatewayURL) => {
|
||||
if (input.baseURL) return input.baseURL
|
||||
if (!input.accountId) throw new Error("CloudflareAIGateway.configure requires accountId unless baseURL is supplied")
|
||||
return `https://gateway.ai.cloudflare.com/v1/${encodeURIComponent(input.accountId)}/${encodeURIComponent(input.gatewayId?.trim() || "default")}/compat`
|
||||
}
|
||||
|
||||
const aiGatewayAuth = (input: AIGatewayOptions) => {
|
||||
if ("auth" in input && input.auth) return input.auth
|
||||
const gateway = Auth.optional(input.gatewayApiKey, "gatewayApiKey")
|
||||
.orElse(Auth.config("CLOUDFLARE_API_TOKEN"))
|
||||
.orElse(Auth.config("CF_AIG_TOKEN"))
|
||||
.pipe(Auth.bearerHeader("cf-aig-authorization"))
|
||||
if (!("apiKey" in input) || input.apiKey === undefined) return gateway
|
||||
if (input.gatewayApiKey === undefined) return Auth.bearer(input.apiKey)
|
||||
return Auth.bearerHeader("cf-aig-authorization", input.gatewayApiKey).andThen(Auth.bearer(input.apiKey))
|
||||
}
|
||||
|
||||
export const workersAIBaseURL = (input: WorkersAIURL) => {
|
||||
if (input.baseURL) return input.baseURL
|
||||
if (!input.accountId) throw new Error("CloudflareWorkersAI.configure requires accountId unless baseURL is supplied")
|
||||
return `https://api.cloudflare.com/client/v4/accounts/${encodeURIComponent(input.accountId)}/ai/v1`
|
||||
}
|
||||
|
||||
const workersAIAuth = (input: WorkersAIOptions) => {
|
||||
return AuthOptions.bearer(input, workersAIAuthEnvVars)
|
||||
}
|
||||
|
||||
export const aiGatewayRoute = OpenAICompatibleChat.route.with({
|
||||
id: "cloudflare-ai-gateway",
|
||||
provider: aiGatewayID,
|
||||
})
|
||||
|
||||
export const workersAIRoute = OpenAICompatibleChat.route.with({
|
||||
id: "cloudflare-workers-ai",
|
||||
provider: workersAIID,
|
||||
})
|
||||
|
||||
export const routes = [aiGatewayRoute, workersAIRoute]
|
||||
|
||||
const aiGatewayDefaults = (options: AIGatewayOptions) => {
|
||||
const {
|
||||
accountId: _accountId,
|
||||
gatewayId: _gatewayId,
|
||||
apiKey: _apiKey,
|
||||
gatewayApiKey: _gatewayApiKey,
|
||||
baseURL: _baseURL,
|
||||
auth: _auth,
|
||||
...rest
|
||||
} = options
|
||||
return rest
|
||||
}
|
||||
|
||||
const workersAIDefaults = (options: WorkersAIOptions) => {
|
||||
const { accountId: _accountId, apiKey: _apiKey, auth: _auth, baseURL: _baseURL, ...rest } = options
|
||||
return rest
|
||||
}
|
||||
|
||||
const configureAIGateway = (options: AIGatewayOptions) => {
|
||||
const route = aiGatewayRoute.with({
|
||||
...aiGatewayDefaults(options),
|
||||
endpoint: { baseURL: aiGatewayBaseURL(options) },
|
||||
auth: aiGatewayAuth(options),
|
||||
})
|
||||
return {
|
||||
id: aiGatewayID,
|
||||
model: (modelID: string | ModelID) => route.model<OpenAIProviderOptionsInput>({ id: modelID }),
|
||||
configure: configureAIGateway,
|
||||
}
|
||||
}
|
||||
|
||||
const configureWorkersAI = (options: WorkersAIOptions) => {
|
||||
const route = workersAIRoute.with({
|
||||
...workersAIDefaults(options),
|
||||
endpoint: { baseURL: workersAIBaseURL(options) },
|
||||
auth: workersAIAuth(options),
|
||||
})
|
||||
return {
|
||||
id: workersAIID,
|
||||
model: (modelID: string | ModelID) => route.model<OpenAIProviderOptionsInput>({ id: modelID }),
|
||||
configure: configureWorkersAI,
|
||||
}
|
||||
}
|
||||
|
||||
export const CloudflareAIGateway = {
|
||||
id: aiGatewayID,
|
||||
configure: configureAIGateway,
|
||||
}
|
||||
|
||||
export const CloudflareWorkersAI = {
|
||||
id: workersAIID,
|
||||
configure: configureWorkersAI,
|
||||
}
|
||||
@@ -1,83 +0,0 @@
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { OpenAICompatibleChat } from "../protocols/openai-compatible-chat.js"
|
||||
import type { RouteDefaultsInput } from "../route/client.js"
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
import { GoogleVertexShared } from "./google-vertex-shared.js"
|
||||
import type { OpenAIProviderOptionsInput } from "./openai-options.js"
|
||||
|
||||
export const id = ProviderID.make("google-vertex")
|
||||
|
||||
export type Config = RouteDefaultsInput &
|
||||
GoogleVertexShared.OAuthOptions & {
|
||||
readonly baseURL?: string
|
||||
readonly location?: string
|
||||
readonly project?: string
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
export interface Settings extends ProviderPackage.Settings {
|
||||
readonly accessToken?: string
|
||||
readonly apiKey?: never
|
||||
readonly baseURL?: string
|
||||
readonly location?: string
|
||||
readonly project?: string
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
const route = OpenAICompatibleChat.route.with({
|
||||
id: "google-vertex-chat",
|
||||
provider: id,
|
||||
})
|
||||
|
||||
export const routes = [route]
|
||||
|
||||
const configuredRoute = (input: Config) => {
|
||||
if ("apiKey" in input && input.apiKey !== undefined) throw new Error("Google Vertex Chat does not support API keys")
|
||||
const {
|
||||
accessToken: _accessToken,
|
||||
auth: _auth,
|
||||
baseURL,
|
||||
location: inputLocation,
|
||||
project: inputProject,
|
||||
...rest
|
||||
} = input
|
||||
const location = GoogleVertexShared.location(inputLocation, "global")
|
||||
const project = GoogleVertexShared.project(inputProject)
|
||||
return route.with({
|
||||
...rest,
|
||||
endpoint: {
|
||||
baseURL:
|
||||
baseURL ??
|
||||
`https://aiplatform.googleapis.com/v1/projects/${GoogleVertexShared.requireProject(project)}/locations/${location}/endpoints/openapi`,
|
||||
},
|
||||
auth: GoogleVertexShared.oauth(input, project),
|
||||
})
|
||||
}
|
||||
|
||||
export const configure = (input: Config = {}) => {
|
||||
const route = configuredRoute(input)
|
||||
return {
|
||||
id,
|
||||
model: (modelID: string | ModelID) => route.model<OpenAIProviderOptionsInput>({ id: modelID }),
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = {
|
||||
id,
|
||||
configure,
|
||||
}
|
||||
|
||||
export const model: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (modelID, settings) => {
|
||||
if (settings.apiKey !== undefined) throw new Error("Google Vertex Chat does not support API keys")
|
||||
return configure({
|
||||
accessToken: settings.accessToken,
|
||||
baseURL: settings.baseURL,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
location: settings.location,
|
||||
project: settings.project,
|
||||
providerOptions: settings.providerOptions,
|
||||
}).model(modelID)
|
||||
}
|
||||
@@ -1,118 +0,0 @@
|
||||
import { Effect, Schema, Struct } from "effect"
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { AnthropicMessages } from "../protocols/anthropic-messages.js"
|
||||
import { Auth } from "../route/auth.js"
|
||||
import { Route, type RouteDefaultsInput } from "../route/client.js"
|
||||
import { Endpoint } from "../route/endpoint.js"
|
||||
import { Framing } from "../route/framing.js"
|
||||
import { Protocol } from "../route/protocol.js"
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
import { GoogleVertexShared } from "./google-vertex-shared.js"
|
||||
|
||||
export type AnthropicOptionsInput = AnthropicMessages.OptionsInput
|
||||
export type AnthropicProviderOptionsInput = AnthropicMessages.ProviderOptionsInput
|
||||
export type AnthropicThinkingInput = AnthropicMessages.ThinkingInput
|
||||
|
||||
const VERSION = "vertex-2023-10-16" as const
|
||||
|
||||
export const id = ProviderID.make("google-vertex")
|
||||
|
||||
export type Config = RouteDefaultsInput &
|
||||
GoogleVertexShared.OAuthOptions & {
|
||||
readonly baseURL?: string
|
||||
readonly location?: string
|
||||
readonly project?: string
|
||||
readonly providerOptions?: AnthropicMessages.ProviderOptionsInput
|
||||
}
|
||||
|
||||
export interface Settings extends ProviderPackage.Settings {
|
||||
readonly accessToken?: string
|
||||
readonly apiKey?: never
|
||||
readonly baseURL?: string
|
||||
readonly location?: string
|
||||
readonly project?: string
|
||||
readonly providerOptions?: AnthropicMessages.ProviderOptionsInput
|
||||
}
|
||||
|
||||
const route = Route.make({
|
||||
id: "google-vertex-messages",
|
||||
provider: id,
|
||||
providerMetadataKey: "anthropic",
|
||||
protocol: Protocol.make({
|
||||
id: AnthropicMessages.protocol.id,
|
||||
body: {
|
||||
schema: Schema.Struct({
|
||||
...Struct.omit(AnthropicMessages.AnthropicMessagesBody.fields, ["model"]),
|
||||
anthropic_version: Schema.Literal(VERSION),
|
||||
}),
|
||||
from: (request) =>
|
||||
AnthropicMessages.protocol.body.from(request).pipe(
|
||||
Effect.map((body) => ({
|
||||
...Struct.omit(body, ["model"]),
|
||||
anthropic_version: VERSION,
|
||||
})),
|
||||
),
|
||||
},
|
||||
stream: AnthropicMessages.protocol.stream,
|
||||
}),
|
||||
endpoint: Endpoint.path(({ request }) => `/${request.model.id}:streamRawPredict`),
|
||||
auth: Auth.none,
|
||||
framing: Framing.sse,
|
||||
})
|
||||
|
||||
export const routes = [route]
|
||||
|
||||
const configuredRoute = (input: Config) => {
|
||||
if ("apiKey" in input && input.apiKey !== undefined)
|
||||
throw new Error("Google Vertex Messages does not support API keys")
|
||||
const {
|
||||
accessToken: _accessToken,
|
||||
auth: _auth,
|
||||
baseURL,
|
||||
location: inputLocation,
|
||||
project: inputProject,
|
||||
...rest
|
||||
} = input
|
||||
const location = GoogleVertexShared.location(inputLocation, "global")
|
||||
const project = GoogleVertexShared.project(inputProject)
|
||||
return route.with({
|
||||
...rest,
|
||||
endpoint: {
|
||||
baseURL:
|
||||
baseURL ??
|
||||
`https://${GoogleVertexShared.host(location)}/v1/projects/${GoogleVertexShared.requireProject(project)}/locations/${location}/publishers/anthropic/models`,
|
||||
},
|
||||
auth: GoogleVertexShared.oauth(input, project),
|
||||
})
|
||||
}
|
||||
|
||||
export const configure = (input: Config = {}) => {
|
||||
const route = configuredRoute(input)
|
||||
return {
|
||||
id,
|
||||
model: (modelID: string | ModelID) => route.model<AnthropicMessages.ProviderOptionsInput>({ id: modelID }),
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = {
|
||||
id,
|
||||
configure,
|
||||
}
|
||||
|
||||
export const model: ProviderPackage.Definition<Settings, AnthropicMessages.ProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) => {
|
||||
if (settings.apiKey !== undefined) throw new Error("Google Vertex Messages does not support API keys")
|
||||
return configure({
|
||||
accessToken: settings.accessToken,
|
||||
baseURL: settings.baseURL,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
location: settings.location,
|
||||
project: settings.project,
|
||||
providerOptions: settings.providerOptions,
|
||||
}).model(modelID)
|
||||
}
|
||||
@@ -1,88 +0,0 @@
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { OpenAICompatibleResponses } from "../protocols/openai-compatible-responses.js"
|
||||
import type { RouteDefaultsInput } from "../route/client.js"
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
import { GoogleVertexShared } from "./google-vertex-shared.js"
|
||||
import type { OpenResponsesProviderOptionsInput } from "./open-responses-options.js"
|
||||
|
||||
export const id = ProviderID.make("google-vertex")
|
||||
|
||||
export type Config = RouteDefaultsInput &
|
||||
GoogleVertexShared.OAuthOptions & {
|
||||
readonly baseURL?: string
|
||||
readonly location?: string
|
||||
readonly project?: string
|
||||
readonly providerOptions?: OpenResponsesProviderOptionsInput
|
||||
}
|
||||
|
||||
export interface Settings extends ProviderPackage.Settings {
|
||||
readonly accessToken?: string
|
||||
readonly apiKey?: never
|
||||
readonly baseURL?: string
|
||||
readonly location?: string
|
||||
readonly project?: string
|
||||
readonly providerOptions?: OpenResponsesProviderOptionsInput
|
||||
}
|
||||
|
||||
const route = OpenAICompatibleResponses.route.with({
|
||||
id: "google-vertex-responses",
|
||||
provider: id,
|
||||
providerOptions: { openresponses: { store: false } },
|
||||
})
|
||||
|
||||
export const routes = [route]
|
||||
|
||||
const configuredRoute = (input: Config) => {
|
||||
if ("apiKey" in input && input.apiKey !== undefined)
|
||||
throw new Error("Google Vertex Responses does not support API keys")
|
||||
const {
|
||||
accessToken: _accessToken,
|
||||
auth: _auth,
|
||||
baseURL,
|
||||
location: inputLocation,
|
||||
project: inputProject,
|
||||
...rest
|
||||
} = input
|
||||
const location = GoogleVertexShared.location(inputLocation, "global")
|
||||
const project = GoogleVertexShared.project(inputProject)
|
||||
return route.with({
|
||||
...rest,
|
||||
endpoint: {
|
||||
baseURL:
|
||||
baseURL ??
|
||||
`https://aiplatform.googleapis.com/v1/projects/${GoogleVertexShared.requireProject(project)}/locations/${location}/endpoints/openapi`,
|
||||
},
|
||||
auth: GoogleVertexShared.oauth(input, project),
|
||||
})
|
||||
}
|
||||
|
||||
export const configure = (input: Config = {}) => {
|
||||
const route = configuredRoute(input)
|
||||
return {
|
||||
id,
|
||||
model: (modelID: string | ModelID) => route.model<OpenResponsesProviderOptionsInput>({ id: modelID }),
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = {
|
||||
id,
|
||||
configure,
|
||||
}
|
||||
|
||||
export const model: ProviderPackage.Definition<Settings, OpenResponsesProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) => {
|
||||
if (settings.apiKey !== undefined) throw new Error("Google Vertex Responses does not support API keys")
|
||||
return configure({
|
||||
accessToken: settings.accessToken,
|
||||
baseURL: settings.baseURL,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
location: settings.location,
|
||||
project: settings.project,
|
||||
providerOptions: settings.providerOptions,
|
||||
}).model(modelID)
|
||||
}
|
||||
@@ -1,77 +0,0 @@
|
||||
import type { AnyAuthClient } from "google-auth-library"
|
||||
import { Effect, Redacted } from "effect"
|
||||
import { Auth, MissingCredentialError } from "../route/auth.js"
|
||||
|
||||
const SCOPE = "https://www.googleapis.com/auth/cloud-platform"
|
||||
|
||||
export type OAuthOptions =
|
||||
| { readonly accessToken?: string; readonly auth?: never }
|
||||
| { readonly accessToken?: never; readonly auth?: Auth.Definition }
|
||||
|
||||
export type ApiKeyOptions =
|
||||
| (OAuthOptions & { readonly apiKey?: never })
|
||||
| { readonly accessToken?: never; readonly apiKey?: string; readonly auth?: never }
|
||||
|
||||
export const project = (value?: string) =>
|
||||
value ??
|
||||
process.env.GOOGLE_VERTEX_PROJECT ??
|
||||
process.env.GOOGLE_CLOUD_PROJECT ??
|
||||
process.env.GCP_PROJECT ??
|
||||
process.env.GCLOUD_PROJECT
|
||||
|
||||
export const location = (value: string | undefined, fallback: string) =>
|
||||
value ??
|
||||
process.env.GOOGLE_VERTEX_LOCATION ??
|
||||
process.env.GOOGLE_CLOUD_LOCATION ??
|
||||
process.env.VERTEX_LOCATION ??
|
||||
fallback
|
||||
|
||||
export const host = (location: string) => {
|
||||
if (location === "global") return "aiplatform.googleapis.com"
|
||||
// Jurisdictional multi-regions use Regional Endpoint Platform domains.
|
||||
if (location === "eu" || location === "us") return `aiplatform.${location}.rep.googleapis.com`
|
||||
return `${location}-aiplatform.googleapis.com`
|
||||
}
|
||||
|
||||
export const requireProject = (value: string | undefined) => {
|
||||
if (value) return value
|
||||
throw new Error("Google Vertex requires a project when baseURL is not configured")
|
||||
}
|
||||
|
||||
export const apiKey = (input: ApiKeyOptions) => {
|
||||
if (input.apiKey !== undefined && (input.accessToken !== undefined || input.auth !== undefined))
|
||||
throw new Error("Google Vertex apiKey cannot be combined with accessToken or auth")
|
||||
if (input.accessToken !== undefined || input.auth !== undefined) return undefined
|
||||
return input.apiKey ?? process.env.GOOGLE_VERTEX_API_KEY
|
||||
}
|
||||
|
||||
const adc = (project?: string) => {
|
||||
let client: Promise<AnyAuthClient> | undefined
|
||||
const loadClient = () => {
|
||||
if (client) return client
|
||||
client = import("google-auth-library").then(({ GoogleAuth }) =>
|
||||
new GoogleAuth({ projectId: project, scopes: [SCOPE] }).getClient(),
|
||||
)
|
||||
return client
|
||||
}
|
||||
return Auth.effect(
|
||||
Effect.tryPromise({
|
||||
try: async () => {
|
||||
const token = await (await loadClient()).getAccessToken()
|
||||
if (!token.token) throw new Error("Google ADC returned an empty access token")
|
||||
return Redacted.make(token.token)
|
||||
},
|
||||
catch: () => new MissingCredentialError("Google Application Default Credentials"),
|
||||
}),
|
||||
).bearer()
|
||||
}
|
||||
|
||||
export const oauth = (input: OAuthOptions, project?: string) => {
|
||||
if (input.accessToken !== undefined && input.auth !== undefined)
|
||||
throw new Error("Google Vertex accessToken cannot be combined with auth")
|
||||
if (input.auth) return input.auth
|
||||
if (input.accessToken !== undefined) return Auth.bearer(input.accessToken)
|
||||
return adc(project)
|
||||
}
|
||||
|
||||
export * as GoogleVertexShared from "./google-vertex-shared.js"
|
||||
@@ -1,106 +0,0 @@
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { Gemini } from "../protocols/gemini.js"
|
||||
import { Auth } from "../route/auth.js"
|
||||
import { Route, type RouteDefaultsInput } from "../route/client.js"
|
||||
import { Endpoint } from "../route/endpoint.js"
|
||||
import { Framing } from "../route/framing.js"
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
import { GoogleVertexShared } from "./google-vertex-shared.js"
|
||||
|
||||
export type GeminiOptionsInput = Gemini.OptionsInput
|
||||
export type GeminiProviderOptionsInput = Gemini.ProviderOptionsInput
|
||||
|
||||
export const id = ProviderID.make("google-vertex")
|
||||
|
||||
export type Config = RouteDefaultsInput &
|
||||
GoogleVertexShared.ApiKeyOptions & {
|
||||
readonly baseURL?: string
|
||||
readonly location?: string
|
||||
readonly project?: string
|
||||
readonly providerOptions?: Gemini.ProviderOptionsInput
|
||||
}
|
||||
|
||||
export type Settings = ProviderPackage.Settings &
|
||||
(
|
||||
| { readonly accessToken?: string; readonly apiKey?: never }
|
||||
| { readonly accessToken?: never; readonly apiKey?: string }
|
||||
) & {
|
||||
readonly baseURL?: string
|
||||
readonly location?: string
|
||||
readonly project?: string
|
||||
readonly providerOptions?: Gemini.ProviderOptionsInput
|
||||
}
|
||||
|
||||
const route = Route.make({
|
||||
id: "google-vertex-gemini",
|
||||
provider: id,
|
||||
providerMetadataKey: "google",
|
||||
protocol: Gemini.protocol,
|
||||
endpoint: Endpoint.path(({ request }) => {
|
||||
const model = String(request.model.id)
|
||||
return `/${model.startsWith("endpoints/") ? model : `models/${model}`}:streamGenerateContent?alt=sse`
|
||||
}),
|
||||
auth: Auth.none,
|
||||
framing: Framing.sse,
|
||||
})
|
||||
|
||||
export const routes = [route]
|
||||
|
||||
const configuredRoute = (input: Config, modelID: string | ModelID) => {
|
||||
const {
|
||||
accessToken: _accessToken,
|
||||
apiKey: _apiKey,
|
||||
auth: _auth,
|
||||
baseURL,
|
||||
location: inputLocation,
|
||||
project: inputProject,
|
||||
...rest
|
||||
} = input
|
||||
const apiKey = GoogleVertexShared.apiKey(input)
|
||||
const endpointModel = String(modelID).startsWith("endpoints/")
|
||||
if (apiKey !== undefined && endpointModel)
|
||||
throw new Error("Google Vertex tuned models do not support Express Mode API keys")
|
||||
const location = GoogleVertexShared.location(inputLocation, "us-central1")
|
||||
const project = GoogleVertexShared.project(inputProject)
|
||||
const endpoint =
|
||||
baseURL ??
|
||||
(apiKey
|
||||
? "https://aiplatform.googleapis.com/v1/publishers/google"
|
||||
: `https://${GoogleVertexShared.host(location)}/v1beta1/projects/${GoogleVertexShared.requireProject(project)}/locations/${location}${endpointModel ? "" : "/publishers/google"}`)
|
||||
return route.with({
|
||||
...rest,
|
||||
endpoint: { baseURL: endpoint },
|
||||
auth: apiKey === undefined ? GoogleVertexShared.oauth(input, project) : Auth.header("x-goog-api-key", apiKey),
|
||||
})
|
||||
}
|
||||
|
||||
export const configure = (input: Config = {}) => {
|
||||
return {
|
||||
id,
|
||||
model: (modelID: string | ModelID) =>
|
||||
configuredRoute(input, modelID).model<Gemini.ProviderOptionsInput>({ id: modelID }),
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = {
|
||||
id,
|
||||
configure,
|
||||
}
|
||||
export const model: ProviderPackage.Definition<Settings, Gemini.ProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) => {
|
||||
if (settings.apiKey !== undefined && settings.accessToken !== undefined)
|
||||
throw new Error("Google Vertex apiKey cannot be combined with accessToken or auth")
|
||||
return configure({
|
||||
...(settings.apiKey === undefined ? { accessToken: settings.accessToken } : { apiKey: settings.apiKey }),
|
||||
baseURL: settings.baseURL,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
location: settings.location,
|
||||
project: settings.project,
|
||||
providerOptions: settings.providerOptions,
|
||||
}).model(modelID)
|
||||
}
|
||||
@@ -1,2 +0,0 @@
|
||||
export { model } from "../google-vertex-chat.js"
|
||||
export type { Settings } from "../google-vertex-chat.js"
|
||||
@@ -1,2 +0,0 @@
|
||||
export { model } from "../google-vertex.js"
|
||||
export type { Settings } from "../google-vertex.js"
|
||||
@@ -1,2 +0,0 @@
|
||||
export { model } from "../google-vertex-messages.js"
|
||||
export type { Settings } from "../google-vertex-messages.js"
|
||||
@@ -1,2 +0,0 @@
|
||||
export { model } from "../google-vertex-responses.js"
|
||||
export type { Settings } from "../google-vertex-responses.js"
|
||||
@@ -1,70 +0,0 @@
|
||||
import type { RouteDefaultsInput } from "../route/client.js"
|
||||
import { Auth } from "../route/auth.js"
|
||||
import type { ProviderAuthOption } from "../route/auth-options.js"
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { HttpOptions, ProviderID, mergeHttpOptions, type ModelID } from "../schema/index.js"
|
||||
import { Gemini } from "../protocols/gemini.js"
|
||||
import { GoogleImages } from "../protocols/google-images.js"
|
||||
|
||||
export type { GoogleImageOptions } from "../protocols/google-images.js"
|
||||
export type GeminiOptionsInput = Gemini.OptionsInput
|
||||
export type GeminiProviderOptionsInput = Gemini.ProviderOptionsInput
|
||||
|
||||
export const id = ProviderID.make("google")
|
||||
|
||||
export const routes = [Gemini.route]
|
||||
|
||||
export type Config = RouteDefaultsInput &
|
||||
ProviderAuthOption<"optional"> & {
|
||||
readonly baseURL?: string
|
||||
readonly providerOptions?: Gemini.ProviderOptionsInput
|
||||
}
|
||||
|
||||
export interface Settings extends ProviderPackage.Settings {
|
||||
readonly apiKey?: string
|
||||
readonly baseURL?: string
|
||||
readonly providerOptions?: Gemini.ProviderOptionsInput
|
||||
}
|
||||
|
||||
const auth = (options: ProviderAuthOption<"optional">) => {
|
||||
if ("auth" in options && options.auth) return options.auth
|
||||
return Auth.optional("apiKey" in options ? options.apiKey : undefined, "apiKey")
|
||||
.orElse(Auth.config("GOOGLE_GENERATIVE_AI_API_KEY"))
|
||||
.pipe(Auth.header("x-goog-api-key"))
|
||||
}
|
||||
|
||||
const configuredRoute = (input: Config) => {
|
||||
const { apiKey: _, auth: _auth, baseURL, ...rest } = input
|
||||
return Gemini.route.with({ ...rest, endpoint: { baseURL }, auth: auth(input) })
|
||||
}
|
||||
|
||||
export const configure = (input: Config = {}) => {
|
||||
const route = configuredRoute(input)
|
||||
const image = (modelID: string | ModelID) =>
|
||||
GoogleImages.model({
|
||||
id: modelID,
|
||||
auth: auth(input),
|
||||
baseURL: input.baseURL,
|
||||
headers: input.headers,
|
||||
http: mergeHttpOptions(input.http === undefined ? undefined : HttpOptions.make(input.http)),
|
||||
})
|
||||
return {
|
||||
id,
|
||||
model: (modelID: string | ModelID) => route.model<Gemini.ProviderOptionsInput>({ id: modelID }),
|
||||
image,
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = configure()
|
||||
export const model: ProviderPackage.Definition<Settings, Gemini.ProviderOptionsInput>["model"] = (modelID, settings) =>
|
||||
configure({
|
||||
apiKey: settings.apiKey,
|
||||
baseURL: settings.baseURL,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
providerOptions: settings.providerOptions,
|
||||
}).model(modelID)
|
||||
|
||||
export const image = provider.image
|
||||
@@ -1,18 +0,0 @@
|
||||
export * as Anthropic from "./anthropic.js"
|
||||
export * as AnthropicCompatible from "./anthropic-compatible.js"
|
||||
export * as AmazonBedrock from "./amazon-bedrock.js"
|
||||
export * as AmazonBedrockMantle from "./amazon-bedrock-mantle.js"
|
||||
export * as Azure from "./azure.js"
|
||||
export * as Cloudflare from "./cloudflare.js"
|
||||
export { CloudflareAIGateway, CloudflareWorkersAI } from "./cloudflare.js"
|
||||
export * as Google from "./google.js"
|
||||
export * as GoogleVertex from "./google-vertex.js"
|
||||
export * as GoogleVertexChat from "./google-vertex-chat.js"
|
||||
export * as GoogleVertexMessages from "./google-vertex-messages.js"
|
||||
export * as GoogleVertexResponses from "./google-vertex-responses.js"
|
||||
export * as OpenAI from "./openai.js"
|
||||
export * as OpenAICompatible from "./openai-compatible.js"
|
||||
export * as OpenAICompatibleResponses from "./openai-compatible-responses.js"
|
||||
export * as OpenRouter from "./openrouter.js"
|
||||
export * as XAI from "./xai.js"
|
||||
export * as ZAI from "./zai.js"
|
||||
@@ -1,19 +0,0 @@
|
||||
import type { ResponseIncludable, ServiceTier } from "../protocols/utils/open-responses-options.js"
|
||||
import type { ProviderOptions, ReasoningEffort, TextVerbosity } from "../schema/index.js"
|
||||
|
||||
export interface OpenResponsesOptionsInput {
|
||||
readonly [key: string]: unknown
|
||||
readonly instructions?: string
|
||||
readonly store?: boolean
|
||||
readonly reasoningEffort?: ReasoningEffort
|
||||
readonly reasoningSummary?: "auto" | "concise" | "detailed"
|
||||
readonly include?: ReadonlyArray<ResponseIncludable>
|
||||
readonly textVerbosity?: TextVerbosity
|
||||
readonly serviceTier?: ServiceTier
|
||||
}
|
||||
|
||||
export type OpenResponsesProviderOptionsInput = ProviderOptions & {
|
||||
readonly openresponses?: OpenResponsesOptionsInput
|
||||
}
|
||||
|
||||
export * as OpenResponsesProviderOptions from "./open-responses-options.js"
|
||||
@@ -1,61 +0,0 @@
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { OpenAICompatibleResponses } from "../protocols/openai-compatible-responses.js"
|
||||
import { AuthOptions, type ProviderAuthOption } from "../route/auth-options.js"
|
||||
import type { RouteDefaultsInput } from "../route/client.js"
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
import type { OpenResponsesProviderOptionsInput } from "./open-responses-options.js"
|
||||
|
||||
export type { OpenResponsesOptionsInput, OpenResponsesProviderOptionsInput } from "./open-responses-options.js"
|
||||
|
||||
export const id = ProviderID.make("openai-compatible")
|
||||
|
||||
export type Config = RouteDefaultsInput &
|
||||
ProviderAuthOption<"optional"> & {
|
||||
readonly provider?: string
|
||||
readonly baseURL: string
|
||||
readonly providerOptions?: OpenResponsesProviderOptionsInput
|
||||
}
|
||||
|
||||
export interface Settings extends ProviderPackage.Settings {
|
||||
readonly apiKey?: string
|
||||
readonly baseURL: string
|
||||
readonly provider?: string
|
||||
readonly providerOptions?: OpenResponsesProviderOptionsInput
|
||||
}
|
||||
|
||||
export const routes = [OpenAICompatibleResponses.route]
|
||||
|
||||
export const configure = (input: Config) => {
|
||||
const provider = input.provider ?? "openai-compatible"
|
||||
const { provider: _, baseURL, apiKey: _apiKey, auth: _auth, ...rest } = input
|
||||
const route = OpenAICompatibleResponses.route.with({
|
||||
...rest,
|
||||
provider,
|
||||
endpoint: { baseURL },
|
||||
auth: AuthOptions.bearer(input, []),
|
||||
})
|
||||
return {
|
||||
id: ProviderID.make(provider),
|
||||
model: (modelID: string | ModelID) => route.model<OpenResponsesProviderOptionsInput>({ id: modelID }),
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = {
|
||||
id,
|
||||
configure,
|
||||
}
|
||||
|
||||
export const model: ProviderPackage.Definition<Settings, OpenResponsesProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) =>
|
||||
configure({
|
||||
apiKey: settings.apiKey,
|
||||
baseURL: settings.baseURL,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
provider: settings.provider,
|
||||
providerOptions: settings.providerOptions,
|
||||
}).model(modelID)
|
||||
@@ -1,86 +0,0 @@
|
||||
import { ProviderID, type ModelID } from "../schema/index.js"
|
||||
import * as OpenAICompatibleChat from "../protocols/openai-compatible-chat.js"
|
||||
import type { RouteDefaultsInput } from "../route/client.js"
|
||||
import { AuthOptions, type ProviderAuthOption } from "../route/auth-options.js"
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { profiles, type OpenAICompatibleProfile } from "./openai-compatible-profile.js"
|
||||
import type { OpenAIProviderOptionsInput } from "./openai-options.js"
|
||||
|
||||
export const id = ProviderID.make("openai-compatible")
|
||||
|
||||
type GenericModelOptions = Omit<RouteDefaultsInput, "providerOptions"> &
|
||||
ProviderAuthOption<"optional"> & {
|
||||
readonly provider?: string
|
||||
readonly baseURL: string
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
export interface Settings extends ProviderPackage.Settings {
|
||||
readonly apiKey?: string
|
||||
readonly baseURL: string
|
||||
readonly provider?: string
|
||||
}
|
||||
|
||||
export type FamilyModelOptions = Omit<RouteDefaultsInput, "providerOptions"> &
|
||||
ProviderAuthOption<"optional"> & {
|
||||
readonly baseURL?: string
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
export const routes = [OpenAICompatibleChat.route]
|
||||
|
||||
export const configure = (input: GenericModelOptions) => {
|
||||
const provider = input.provider ?? "openai-compatible"
|
||||
const { provider: _, baseURL, apiKey: _apiKey, auth: _auth, ...rest } = input
|
||||
const route = OpenAICompatibleChat.route.with({
|
||||
...rest,
|
||||
provider,
|
||||
endpoint: { baseURL },
|
||||
auth: AuthOptions.bearer(input, []),
|
||||
})
|
||||
return {
|
||||
id: ProviderID.make(provider),
|
||||
model: (modelID: string | ModelID) =>
|
||||
route.model<OpenAIProviderOptionsInput>({ id: modelID, provider: ProviderID.make(provider) }),
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
const define = (profile: OpenAICompatibleProfile) => {
|
||||
const configureProfile = (input: FamilyModelOptions = {}) => {
|
||||
const facade = configure({
|
||||
...input,
|
||||
baseURL: input.baseURL ?? profile.baseURL,
|
||||
provider: profile.provider,
|
||||
})
|
||||
return {
|
||||
id: ProviderID.make(profile.provider),
|
||||
model: facade.model,
|
||||
configure: configureProfile,
|
||||
}
|
||||
}
|
||||
return configureProfile()
|
||||
}
|
||||
|
||||
export const provider = {
|
||||
id,
|
||||
configure,
|
||||
}
|
||||
|
||||
export const model: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (modelID, settings) =>
|
||||
configure({
|
||||
apiKey: settings.apiKey,
|
||||
baseURL: settings.baseURL,
|
||||
headers: settings.headers === undefined ? undefined : { ...settings.headers },
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
provider: settings.provider,
|
||||
}).model(modelID)
|
||||
|
||||
export const baseten = define(profiles.baseten)
|
||||
export const cerebras = define(profiles.cerebras)
|
||||
export const deepinfra = define(profiles.deepinfra)
|
||||
export const deepseek = define(profiles.deepseek)
|
||||
export const fireworks = define(profiles.fireworks)
|
||||
export const groq = define(profiles.groq)
|
||||
export const togetherai = define(profiles.togetherai)
|
||||
@@ -1 +0,0 @@
|
||||
export * from "../openai-compatible-responses.js"
|
||||
@@ -1,70 +0,0 @@
|
||||
import type { ProviderOptions } from "../schema/index.js"
|
||||
import { mergeProviderOptions } from "../schema/index.js"
|
||||
import type { OpenResponsesOptionsInput } from "./open-responses-options.js"
|
||||
|
||||
export type { OpenAIResponseIncludable, OpenAIServiceTier } from "../protocols/utils/openai-options.js"
|
||||
|
||||
export type OpenAIOptionsInput = OpenResponsesOptionsInput
|
||||
|
||||
export type OpenAIProviderOptionsInput = ProviderOptions & {
|
||||
readonly openai?: OpenAIOptionsInput
|
||||
}
|
||||
|
||||
const definedEntries = (input: Record<string, unknown>) =>
|
||||
Object.entries(input).filter((entry) => entry[1] !== undefined)
|
||||
|
||||
const openAIProviderOptions = (options: OpenAIOptionsInput | undefined): ProviderOptions | undefined => {
|
||||
const openai = Object.fromEntries(
|
||||
definedEntries({
|
||||
store: options?.store,
|
||||
reasoningEffort: options?.reasoningEffort,
|
||||
reasoningSummary: options?.reasoningSummary,
|
||||
include: options?.include,
|
||||
textVerbosity: options?.textVerbosity,
|
||||
serviceTier: options?.serviceTier,
|
||||
}),
|
||||
)
|
||||
if (Object.keys(openai).length === 0) return undefined
|
||||
return { openai }
|
||||
}
|
||||
|
||||
export const gpt5DefaultOptions = (
|
||||
modelID: string,
|
||||
options: { readonly textVerbosity?: boolean } = {},
|
||||
): ProviderOptions | undefined => {
|
||||
const id = modelID.toLowerCase()
|
||||
if (!id.includes("gpt-5") || id.includes("gpt-5-chat") || id.includes("gpt-5-pro")) return undefined
|
||||
return openAIProviderOptions({
|
||||
reasoningEffort: "medium",
|
||||
reasoningSummary: "auto",
|
||||
// GPT-5 reasoning models are configured stateless (`store: false`) by
|
||||
// `openAIDefaultOptions` below, so the only way a follow-up turn can
|
||||
// carry reasoning state is via the encrypted reasoning include. Without
|
||||
// this, callers using the default model facade get reasoning summaries
|
||||
// they cannot replay statelessly.
|
||||
include: ["reasoning.encrypted_content"],
|
||||
textVerbosity:
|
||||
options.textVerbosity === true && id.includes("gpt-5.") && !id.includes("codex") && !id.includes("-chat")
|
||||
? "low"
|
||||
: undefined,
|
||||
})
|
||||
}
|
||||
|
||||
export const openAIDefaultOptions = (
|
||||
modelID: string,
|
||||
options: { readonly textVerbosity?: boolean } = {},
|
||||
): ProviderOptions | undefined =>
|
||||
mergeProviderOptions(openAIProviderOptions({ store: false }), gpt5DefaultOptions(modelID, options))
|
||||
|
||||
export const withOpenAIOptions = <Options extends { readonly providerOptions?: OpenAIProviderOptionsInput }>(
|
||||
modelID: string,
|
||||
options: Options,
|
||||
defaults: { readonly textVerbosity?: boolean } = {},
|
||||
): Omit<Options, "providerOptions"> & { readonly providerOptions?: ProviderOptions } => {
|
||||
return {
|
||||
...options,
|
||||
providerOptions: mergeProviderOptions(openAIDefaultOptions(modelID, defaults), options.providerOptions),
|
||||
}
|
||||
}
|
||||
|
||||
export * as OpenAIProviderOptions from "./openai-options.js"
|
||||
@@ -1,143 +0,0 @@
|
||||
import { AuthOptions, type ProviderAuthOption } from "../route/auth-options.js"
|
||||
import type { Route, RouteDefaultsInput } from "../route/client.js"
|
||||
import type { ProviderPackage } from "../provider-package.js"
|
||||
import { HttpOptions, ProviderID, ToolDefinition, mergeHttpOptions, type ModelID } from "../schema/index.js"
|
||||
import * as OpenAIChat from "../protocols/openai-chat.js"
|
||||
import * as OpenAIResponses from "../protocols/openai-responses.js"
|
||||
import { withOpenAIOptions, type OpenAIProviderOptionsInput } from "./openai-options.js"
|
||||
import { OpenAIImages, type OpenAIImageString } from "../protocols/openai-images.js"
|
||||
|
||||
export type { OpenAIOptionsInput, OpenAIResponseIncludable } from "./openai-options.js"
|
||||
export type { OpenAIImageOptions } from "../protocols/openai-images.js"
|
||||
|
||||
export const id = ProviderID.make("openai")
|
||||
|
||||
export const routes = [OpenAIResponses.route, OpenAIChat.route]
|
||||
|
||||
// This provider facade wraps the lower-level Responses and Chat model factories
|
||||
// with OpenAI-specific conveniences: typed options, API-key sugar, env fallback,
|
||||
// and default option normalization.
|
||||
export type Config = RouteDefaultsInput &
|
||||
ProviderAuthOption<"optional"> & {
|
||||
readonly baseURL?: string
|
||||
readonly queryParams?: Record<string, string>
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
export interface ImageGenerationOptions {
|
||||
readonly action?: OpenAIImageString<"auto" | "generate" | "edit">
|
||||
readonly background?: OpenAIImageString<"auto" | "opaque" | "transparent">
|
||||
readonly inputFidelity?: OpenAIImageString<"low" | "high">
|
||||
readonly outputCompression?: number
|
||||
readonly outputFormat?: OpenAIImageString<"png" | "jpeg" | "webp">
|
||||
readonly partialImages?: number
|
||||
readonly quality?: OpenAIImageString<"auto" | "low" | "medium" | "high" | "standard" | "hd">
|
||||
readonly size?: OpenAIImageString<
|
||||
"auto" | "256x256" | "512x512" | "1024x1024" | "1536x1024" | "1024x1536" | "1792x1024" | "1024x1792"
|
||||
>
|
||||
}
|
||||
|
||||
export const imageGeneration = (options: ImageGenerationOptions = {}) =>
|
||||
ToolDefinition.make({
|
||||
name: "image_generation",
|
||||
description: "Generate or edit an image using OpenAI's hosted image generation tool.",
|
||||
inputSchema: { type: "object", properties: {}, additionalProperties: false },
|
||||
native: {
|
||||
openai: {
|
||||
type: "image_generation",
|
||||
action: options.action,
|
||||
background: options.background,
|
||||
input_fidelity: options.inputFidelity,
|
||||
output_compression: options.outputCompression,
|
||||
output_format: options.outputFormat,
|
||||
partial_images: options.partialImages,
|
||||
quality: options.quality,
|
||||
size: options.size,
|
||||
},
|
||||
},
|
||||
})
|
||||
|
||||
export interface Settings extends ProviderPackage.Settings {
|
||||
readonly apiKey?: string
|
||||
readonly baseURL?: string
|
||||
readonly organization?: string
|
||||
readonly project?: string
|
||||
readonly queryParams?: Readonly<Record<string, string>>
|
||||
readonly providerOptions?: OpenAIProviderOptionsInput
|
||||
}
|
||||
|
||||
const auth = (options: ProviderAuthOption<"optional">) => AuthOptions.bearer(options, "OPENAI_API_KEY")
|
||||
|
||||
const defaults = (input: Config) => {
|
||||
const { apiKey: _, auth: _auth, baseURL: _baseURL, queryParams: _queryParams, ...rest } = input
|
||||
return rest
|
||||
}
|
||||
|
||||
const configuredRoute = <Body, Prepared>(route: Route<Body, Prepared>, input: Config) =>
|
||||
route.with({
|
||||
auth: auth(input),
|
||||
endpoint: { baseURL: input.baseURL, query: input.queryParams },
|
||||
})
|
||||
|
||||
export const configure = (input: Config = {}) => {
|
||||
const responsesRoute = configuredRoute(OpenAIResponses.route, input)
|
||||
const chatRoute = configuredRoute(OpenAIChat.route, input)
|
||||
const modelDefaults = defaults(input)
|
||||
const responses = (id: string | ModelID) =>
|
||||
responsesRoute
|
||||
.with(withOpenAIOptions(id, modelDefaults, { textVerbosity: true }))
|
||||
.model<OpenAIProviderOptionsInput>({ id })
|
||||
const chat = (id: string | ModelID) =>
|
||||
chatRoute.with(withOpenAIOptions(id, modelDefaults)).model<OpenAIProviderOptionsInput>({ id })
|
||||
const image = (modelID: string | ModelID) =>
|
||||
OpenAIImages.model({
|
||||
id: modelID,
|
||||
auth: auth(input),
|
||||
baseURL: input.baseURL,
|
||||
headers: input.headers,
|
||||
http: mergeHttpOptions(
|
||||
input.http === undefined ? undefined : HttpOptions.make(input.http),
|
||||
input.queryParams === undefined ? undefined : new HttpOptions({ query: input.queryParams }),
|
||||
),
|
||||
})
|
||||
|
||||
return {
|
||||
id,
|
||||
model: responses,
|
||||
responses,
|
||||
chat,
|
||||
image,
|
||||
configure,
|
||||
}
|
||||
}
|
||||
|
||||
export const provider = configure()
|
||||
|
||||
const config = (settings: Settings): Config => {
|
||||
const headers = {
|
||||
...(settings.organization === undefined ? {} : { "OpenAI-Organization": settings.organization }),
|
||||
...(settings.project === undefined ? {} : { "OpenAI-Project": settings.project }),
|
||||
...settings.headers,
|
||||
}
|
||||
return {
|
||||
apiKey: settings.apiKey,
|
||||
baseURL: settings.baseURL,
|
||||
headers: Object.keys(headers).length === 0 ? undefined : headers,
|
||||
http: settings.body === undefined ? undefined : { body: { ...settings.body } },
|
||||
limits: settings.limits,
|
||||
providerOptions: settings.providerOptions,
|
||||
queryParams: settings.queryParams === undefined ? undefined : { ...settings.queryParams },
|
||||
}
|
||||
}
|
||||
|
||||
export const model: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (modelID, settings) => {
|
||||
return configure(config(settings)).responses(modelID)
|
||||
}
|
||||
|
||||
export const chatModel: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (
|
||||
modelID,
|
||||
settings,
|
||||
) => configure(config(settings)).chat(modelID)
|
||||
export const responses = provider.responses
|
||||
export const chat = provider.chat
|
||||
export const image = provider.image
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user