feat(core): manage session websocket channels

.gitattributes

@@ -1,4 +1,3 @@
 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

.github/TEAM_MEMBERS

@@ -10,7 +10,6 @@ kitlangton
 kommander
 ludvigrask
 MrMushrooooom
-neriousy
 nexxeln
 R44VC0RP
 rekram1-node

.github/workflows/generate.yml

@@ -4,7 +4,6 @@ on:
   push:
     branches:
       - dev
-      - v2
 
 jobs:
   generate:

.github/workflows/publish.yml

@@ -75,7 +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:
@@ -91,7 +91,7 @@ jobs:
           opencode-app-secret: ${{ secrets.OPENCODE_APP_SECRET }}
 
       - name: Build legacy CLI
-        if: github.ref_name != 'v2' && github.ref_name != 'beta'
+        if: github.ref_name != 'v2'
         run: ./packages/opencode/script/build.ts ${{ (github.ref_name == 'beta' && '--sourcemaps') || '' }}
         env:
           OPENCODE_VERSION: ${{ needs.version.outputs.version }}
@@ -109,7 +109,7 @@ jobs:
           GH_TOKEN: ${{ steps.committer.outputs.token }}
 
       - uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
-        if: github.ref_name != 'v2' && github.ref_name != 'beta'
+        if: github.ref_name != 'v2'
         with:
           name: opencode-cli
           path: |
@@ -117,76 +117,22 @@ jobs:
             packages/opencode/dist/opencode-linux*
 
       - uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
-        if: github.ref_name != 'v2' && github.ref_name != 'beta'
+        if: github.ref_name != 'v2'
         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'
+    if: github.repository == 'anomalyco/opencode'
     strategy:
       fail-fast: false
       matrix:
@@ -238,7 +184,7 @@ jobs:
       - 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 }}
@@ -335,6 +281,7 @@ jobs:
 
   build-electron:
     needs:
+      - build-cli
       - version
     if: github.repository == 'anomalyco/opencode' && github.ref_name != 'v2'
     continue-on-error: false
@@ -431,14 +378,16 @@ 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 }}
@@ -447,7 +396,6 @@ 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
@@ -457,7 +405,8 @@ 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 }}
@@ -522,7 +471,6 @@ jobs:
     needs:
       - version
       - build-cli
-      - sign-cli-macos
       - build-node-cli
       - sign-cli-windows
       - build-electron
@@ -552,31 +500,29 @@ jobs:
           registry-url: "https://registry.npmjs.org"
 
       - 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
           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-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

.opencode/plugins/tui/discovery-smoke.ts

@@ -0,0 +1,13 @@
+import type { Context } from "../../../packages/plugin/src/tui/context"
+
+export default {
+  id: "test.tui-discovery-smoke",
+  setup(_context: Context) {
+    // context.ui.toast.show({
+    //   title: "TUI plugin discovery works",
+    //   message: "Loaded .opencode/plugins/tui/discovery-smoke.ts",
+    //   variant: "success",
+    //   duration: 30_000,
+    // })
+  },
+}

.opencode/skills/ideal-pseudocode/SKILL.md

@@ -0,0 +1,68 @@
+---
+name: ideal-pseudocode
+description: Function-by-function refactoring loop driven by ideal pseudocode. Use when the user says "ideal pseudocode", asks to make a function read like its pseudocode, or wants a dense module cleaned up one function at a time.
+---
+
+# Ideal Pseudocode
+
+Clean up one function at a time by writing the pseudocode it _should_ read as, naming every delta between that and the real code, and closing only the gaps the user approves.
+
+## Loop
+
+One function per round. Never touch code before the user picks a direction.
+
+1. **Pick the target** with the user — usually the next function up or down the call chain from the last round.
+2. **Read the current code** fresh from disk. It may have unsaved or parallel edits; ask before overwriting anything unexpected.
+3. **Distill.** Write the function's ideal pseudocode in a `ts`-fenced code block — TypeScript-flavored for syntax highlighting, but pseudocode: comments over mechanics, one line per idea, every arm of a loop visible as an arm. For a dense or unfamiliar function, first show the _current_ structure as pseudocode, then the ideal.
+4. **Name the deltas.** A numbered list; each delta is one concrete gap: control flow smuggled through combinators, a flag-typed parameter, colliding names, side-effect plumbing drowning a loop body, twin functions differing only by a filter. For each, give an honest cost/benefit and a decisive recommendation — including "leave it" when the density has reasons (state the reasons).
+5. **Verify before proposing.** Any API the proposal builds on (a combinator, schedule, matcher) gets checked against the installed version's source, not memory.
+6. **Get buy-in, then apply.** Minimal edits. Quarantine irreducible ugliness inside a small helper whose name is the pseudocode line it implements. Collapse shallow sibling functions into one deep function only when the distinguishing argument is a domain noun, not a behavior switch.
+7. **Validate.** Typecheck and the focused test suites from the package directory; rerun timing-sensitive tests repeatedly. Flag any semantic change the restructure introduced, even when tests pass.
+8. **Offer the next target** and stop.
+
+## Style
+
+- Keep each round tight: pseudocode block, deltas, recommendation, one question.
+- State mutation belongs in one visible place (usually the loop tail), never spread across error channels or callback closures.
+- Prefer early return and `if` chains for control flow; matchers are for producing values.
+- Compare names against the question each caller is actually asking; rename toward the domain noun.
+
+## Example round
+
+Every round should read like this (abridged from a real one, on an Effect step-runner):
+
+> `runStep` it is. The ideal pseudocode:
+>
+> ```ts
+> // One logical step: one settled model call, however many attempts it takes.
+> // An attempt can end without settling in two ways:
+> //   transient provider failure -> retry the same call, same assistant message
+> //   compaction rewrote history -> rebuild the request and call again
+> function runStep(session, promotable, step) {
+>   while (true) {
+>     const result = callModel(session, promotable, step)
+>
+>     if (result.completed) return result
+>
+>     if (result.retryable) {
+>       wait(retrySchedule) // exhausted -> publish Step.Failed, fail drain
+>       promotable = none // never re-promote on a second attempt
+>       continue
+>     }
+>
+>     // compaction restarted the step: fresh request from rewritten history
+>     promotable = none
+>     step = result.step
+>   }
+> }
+> ```
+>
+> Comparing against the real thing, three deltas:
+>
+> 1. **The retry arm doesn't live in the loop.** In the pseudocode, retry is just another loop arm. In the real code it's smuggled through the error channel: `Effect.tapError` mutating loop variables via `Effect.sync` closures, then `Effect.retryOrElse` re-running a suspended call. Four mutable variables mutated from three different places — the state flow is invisible unless you already know the trick.
+> 2. **`recoverOverflow: typeof compaction.compact | undefined`** — a function-typed parameter used as a one-shot armed/disarmed flag. The callee already closes over `compaction`; a boolean says what it is.
+> 3. **`assistantMessageID`** exists solely to resume the same durable message across retries — fine, but only legible once the retry arm is visible in the loop.
+>
+> The unifying move for (1): a schedule driver, so a retryable failure becomes a caught value handled in the loop body exactly like the compaction arm. One honest trade: it's more manual than the blessed combinator — but the manual version is the loop the combinator was hiding, and the loop already exists for restarts. Two restart mechanisms, one control structure.
+>
+> Want me to apply it — unified loop, simplified schedule input, boolean `recoverOverflow`?

.opencode/skills/opencode-drive/SKILL.md

@@ -251,3 +251,4 @@ opencode-drive stop --name demo
 ```bash
 opencode-drive dir --name demo
 ```
+

.opencode/skills/rtl-aware-development/SKILL.md

@@ -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/)

.opencode/tool/github-triage.ts

@@ -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"],
   inference: ["fwang", "MrMushrooooom", "starptech"],
   windows: ["Hona"],
 } as const

AGENTS.md

@@ -1,9 +1,8 @@
 - 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.
 - 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 `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.
+- The default branch in this repo is `dev`.
+- Local `main` ref may not exist; use `dev` or `origin/dev` for diffs.
 
 ## Live V2 TUI Testing
 

CONTRIBUTING.md

@@ -252,6 +252,33 @@ These are not strictly enforced, they are just general guidelines:
 
 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:

docs/design/v1-v2-database-migration.md

@@ -5,27 +5,8 @@
 - Use the `dev` branch database schema and migration registry as the V1 baseline.
 - Remove migrations that exist only on the V2 branch.
 - Generate one canonical migration from the `dev` schema to the final V2 schema.
-- Keep the canonical migration focused on schema changes and dropping obsolete tables.
-- Run the V1 history backfill through an experimental server endpoint invoked by the CLI before it opens the TUI.
-- Show committed session progress while the endpoint runs.
-
-Expose `GET /api/experimental/migration/v1` for status and a blocking `POST /api/experimental/migration/v1` to run or
-resume the backfill. The status is `required`, `running`, or `completed`. On startup, the CLI checks status first and
-renders no migration UI when it is already complete. For required or running status, it shows a spinner and waits for the
-blocking POST without a request timeout. While migration runs, poll GET once per second and render completed and total
-session counts. GET derives total from all session rows and completed from rows through the stored cursor; the count
-advances only after a session transaction commits. The POST returns `{ status: "completed" }`. Do not add a background
-job or streaming progress protocol. Interrupted calls resume from the stored cursor.
-Initially, only interactive TUI startup performs this check; noninteractive run, ACP, raw API, service, health, version,
-and help flows do not trigger the backfill.
-
-Keep migration behavior in Core: status, semaphore, checkpointing, V1 decoding, transformation, and database writes.
-Protocol owns the experimental GET/POST contracts, Server handlers delegate to Core, and the interactive CLI owns only
-the status check and spinner presentation.
-
-Guard the endpoint with one process-local Effect `Semaphore`. Concurrent callers wait; after the active call completes,
-waiting callers acquire the permit, observe the completion key, and return immediately. No distributed lock is required
-for the current single elected server process.
+- Add explicit data operations to that migration where generated DDL is insufficient.
+- Test the migration against a populated database at the exact `dev` schema.
 
 ## Preserve
 
@@ -34,35 +15,20 @@ The canonical V1 data remains in its existing tables. In particular, preserve `s
 Preserve `workspace` rows and existing `session.workspace_id` values unchanged. The migration must not clear or rebuild
 workspace relationships.
 
-Preserve existing non-null `session.agent` and `session.model` selections. Fill missing values from the latest ordinary
-V1 user message ordered by `time_created` and `id`, excluding compaction and subtask-only messages. Copy agent, provider
-ID, model ID, and variant, normalizing an absent variant to `default`.
+Keep the `todo` table and its data unchanged. V2 does not currently migrate todos into another representation, and the
+generated migration must not drop the table.
 
-Recompute session usage aggregates from all canonical V1 assistant messages, including compaction or other internal
-assistants omitted from the V2 projection. Overwrite session cost and input, output, reasoning, cache-read, and
-cache-write token totals with those sums.
+## Truncate
 
-Clear persisted `session.revert` state. A staged revert is transient operational state and may refer to omitted projection
-rows or unavailable snapshots; it must not resume automatically after upgrading. Preserve the underlying messages,
-parts, and file history.
+Truncate these pre-launch V2 tables before applying schema changes:
 
-Clear `session.time_compacting`, leave the new `time_suspended` column as `NULL`, and preserve session creation, update,
-and archive timestamps. Preserve project `time_initialized`; it is unrelated durable state.
+- `event`
+- `event_sequence`
+- `session_message`
 
-Keep the legacy `todo` table and its data physically unchanged, but do not include it in the final V2 Drizzle schema.
-After generation, remove the generated `DROP TABLE todo` statement from the canonical migration so the table remains as
-unmanaged legacy storage.
-
-## Per-Session Replacement
-
-Do not truncate `event`, `event_sequence`, or `session_message` globally before the backfill. A whole-table delete can
-hold SQLite's writer lock long enough to block the running TUI.
-
-Replace each legacy session's V2 state inside that session's checkpointed migration transaction. Delete `event` rows for
-the session aggregate, delete its `session_message` rows, rebuild its projection from canonical V1 `message` and `part`
-rows, and overwrite its `event_sequence` watermark. If migration of that session fails, all replacements roll back and
-the durable cursor remains at the previously committed session. Rows owned by sessions outside the legacy migration set
-remain untouched.
+These rows are not canonical V1 data. Truncating `event` before adding the required `event.created` column means the
+column needs neither a backfill nor a default. After truncation, rebuild `session_message` from canonical V1 `message`
+and `part` rows rather than retaining its pre-launch V2 contents.
 
 ## Message Backfill
 
@@ -70,23 +36,9 @@ Backfill canonical V1 history from `message` and `part` into `session_message`.
 the migration. Preserving the V1 tables alone keeps the data safe but does not make existing history visible through the
 V2 session APIs, which read `session_message`.
 
-Do not fail the whole migration when a V1 message or part payload cannot be decoded. Skip an undecodable message's V2
-projection and log its session and message IDs. Skip an undecodable part while continuing to map its message, and perform
-special-message pairing only with decoded rows. Assign sequences after filtering. Leave every malformed source row
-untouched in the V1 tables.
-
-Skip and log orphan parts whose source message does not exist and parts with unknown or unsupported types. Continue
-migrating the owning message and other valid parts. Include session, message, part ID, and observed type in warnings, and
-leave skipped source rows unchanged.
-
 Reuse each V1 `message.id` as the corresponding `session_message.id`. Stable IDs keep the migration deterministic and
 avoid rewriting other persisted state that may refer to a message.
 
-For ordinary user and assistant rows, preserve source `message.time_created` and `message.time_updated`. Entirely
-synthetic messages preserve their source timestamps, and synthetic rows split from mixed messages use the source user
-timestamps. A collapsed compaction uses the compaction user creation time and the later update time of the compaction
-user and summary assistant. Keep payload creation/completion times consistent with row timestamps.
-
 Within each session, order V1 messages by `time_created` and then `id`, matching the existing V1 message index. Assign
 contiguous `session_message.seq` values starting at `0`.
 
@@ -94,122 +46,15 @@ Map ordinary V1 messages one-to-one by role. Each ordinary V1 user message becom
 V1 assistant message becomes one V2 `assistant` row. Fold the source message's ordered V1 parts into that row's V2
 payload.
 
-Keep ordinary messages even when their transformed payload becomes empty after filtering. Preserve an empty V2 user row
-with `text: ""` and an empty V2 assistant row with `content: []` so IDs, chronology, and conversation structure remain
-stable. Omit only explicitly dropped internal concepts and undecodable messages.
-
 Handle semantic marker parts before applying the ordinary mapping. In particular, a V1 user message containing a
 `compaction` part and its paired assistant summary represent one compaction operation, not two ordinary messages. Special
 part mappings must be decided explicitly before implementing the backfill.
 
-Do not carry the V1 subtask concept into the V2 projection. Omit user messages containing only `subtask` parts and omit
-the paired assistant task-tool messages generated from those markers. For mixed user messages, ignore the `subtask`
-parts while preserving ordinary content, and still omit assistant task-tool messages generated by the skipped subtasks.
-Keep all source rows unchanged in the V1 `message` and `part` tables.
-
-Map ordinary V1 assistant `text` and `reasoning` parts into the V2 assistant `content` array in part order. Preserve text,
-including empty assistant text parts used as structural separators. Map V1 part metadata to optional V2 provider state.
-For reasoning, map `time.start` to `time.created` and optional `time.end` to `time.completed`.
-
-Preserve V1 tool parts that are `pending` or `running`, but convert them to terminal V2 tool error states. Preserve the
-call ID, tool name, parsed input, metadata, and available start time. Use the assistant message creation time when the V1
-state has no start time. Set the error to type `tool.interrupted` with message
-`Tool execution was interrupted before V2 migration`. Never resume migrated tool executions.
-
-For a completed V1 tool part, use `callID` as the V2 tool content ID and preserve the tool name and parsed input. Set the
-state to `completed`. Convert V1 output into the first text content item and convert stored output attachments into
-following file content items with their URI, MIME type, and filename. Preserve state metadata. Map `time.start` to
-`time.created` and `time.end` to `time.completed`. When `time.compacted` exists, use
-`[Old tool result content cleared]` as the only output and omit attachments.
-
-For a failed V1 tool part, preserve the call ID, tool name, parsed input, metadata, and timestamps, and set the V2 state
-to `error`. Convert the V1 error string to a structured error with type `tool.execution`. If V1 metadata contains a string
-`output`, preserve it as optional V2 text content. Map `time.start` to `time.created` and `time.end` to `time.completed`.
-
-For an ordinary V1 assistant message, preserve agent, provider ID, model ID, optional variant, creation and completion
-times, cost, and input/output/reasoning/cache token counts. Use `default` when the V1 variant is absent. Ignore V1
-`tokens.total` because it is derivable and V2 does not persist it.
-
-Use V1 assistant `parentID` only while pairing compactions and skipped subtasks with their originating user messages. Do
-not persist it in ordinary V2 assistant rows; V2 uses ordered history rather than user/assistant parent links.
-
-Ignore the optional V1 assistant `structured` output value. V2 has no equivalent top-level assistant field, and visible
-text and tool content are migrated separately. Retain the original structured value only in the V1 `message` row.
-
-Ignore V1 assistant `mode` and historical `path` (`cwd` and `root`). Mode is redundant with the preserved assistant
-agent, and historical filesystem paths do not belong to the V2 assistant message contract. Retain them only in the V1
-`message` row.
-
-For assistant finish reasons, preserve `stop`, `length`, `tool-calls`, `content-filter`, `error`, and `unknown`. Map every
-other nonempty V1 finish value to `unknown`, and leave the field absent when V1 omitted it. Do not retain unrecognized raw
-finish values in metadata.
-
-Map V1 assistant errors into the current V2 `{ type, message }` storage shape. Normalize Auth, content-filter, context
-overflow, structured-output, output-length, aborted, API, and unknown errors to the established V2 string conventions,
-preserve the message, and discard V1-only retryability and raw provider details.
-
-Ignore V1 `retry` parts. Do not populate the V2 assistant `retry` field during migration; historical retry state is not
-useful enough to preserve. The original retry rows remain in the V1 `part` table.
-
-Do not emit V2 assistant content for V1 `step-start` and `step-finish` parts. Use the first available
-`step-start.snapshot` as `assistant.snapshot.start` and the last available `step-finish.snapshot` as
-`assistant.snapshot.end`. Continue to source finish, cost, and tokens from the assistant message itself. Ignore step
-markers without snapshots.
-
-Do not emit assistant content for standalone V1 `snapshot` or `patch` parts. If no start snapshot came from `step-start`,
-use the first standalone snapshot value, then the first patch hash as a final fallback. Only `step-finish.snapshot` may
-populate the end snapshot. Merge patch file lists into `assistant.snapshot.files` in first-seen order with duplicates
-removed.
-
-V2 follow-up: replace the open `SessionError.Error` string shape with a properly typed persisted error union. This is not
-a blocker for the V1 migration, which should target the current storage contract.
-
 V1 synthetic content is represented by user text parts with `synthetic: true`, not by a separate message role. A V1 user
 message whose visible text parts are all synthetic should become a V2 `synthetic` message. If a V1 user message mixes
 ordinary and synthetic content, preserve the ordinary content in the V2 `user` row and emit the synthetic content as an
 adjacent V2 `synthetic` row. Ignore text parts marked `ignored`, matching V1 model-history behavior.
 
-For an ordinary V2 user message, take visible V1 text parts that are neither ignored nor synthetic, preserve part order,
-and join their text with `"\n\n"`. Use an empty string when the message contains attachments but no ordinary text.
-
-Ignore the optional V1 user-message `system` override. Do not create a V2 system message or preserve the override in
-metadata. The original value remains in the V1 `message` row.
-
-Ignore the optional V1 user-message `tools` map. It represented request-time tool enablement for a historical step and
-must not affect future V2 execution. The original value remains in the V1 `message` row.
-
-Ignore the optional V1 user-message `format` field and its schema. It controlled structured-output behavior for a
-historical request and must not affect future V2 runs. Preserve visible assistant text normally; retain the original
-format only in the V1 `message` row.
-
-Ignore V1 user-message `summary` metadata, including title, body, and diffs. V2 user messages have no equivalent field,
-and session-level summary data is already persisted separately. Retain the original summary only in the V1 `message`
-row.
-
-Map V1 `agent` parts into the V2 user message's `agents` array in part order. Preserve `name`. When the V1 part has
-`source`, map its `value`, `start`, and `end` into the V2 attachment's `mention.text`, `mention.start`, and `mention.end`.
-Omit `agents` when there are no agent parts.
-
-Do not read the filesystem or network while migrating V1 file attachments. Attachment migration must be deterministic
-from database contents alone. Convert persisted `data:` URLs; represent non-embedded `file:`, HTTP, and other external
-URLs with deterministic text rather than fetching them. Keep the original V1 `part` rows unchanged.
-
-For a V1 file backed by a `data:` URL, decode the URL and normalize its payload to base64 for the V2 attachment's `data`.
-Preserve `mime` and optional `filename` as `name`. Use a V2 `uri` source with the original URI for a V1 resource source;
-otherwise use an `inline` source. When V1 source text metadata exists, map its `value`, `start`, and `end` into the V2
-attachment mention. Leave `description` unset and preserve file-part order in the V2 `files` array.
-
-For a non-embedded V1 file, do not create a V2 file attachment. Append
-`[Attachment unavailable after migration: <name-or-url> (<mime>)]` to the V2 user text in original part order, separated
-by blank lines. Prefer the V1 filename, then resource URI, then part URL for the label. The original URL remains only in
-the preserved V1 `part` row.
-
-For a synthetic row split from a mixed user message, derive a generated-looking ID from the source message ID. Preserve
-the source ID's 12-character timestamp component and replace its 14-character random component with a deterministic
-base-62 encoding of a hash of `v1-synthetic:` plus the source message ID. If that candidate collides with an existing or
-derived message ID, deterministically retry with an incrementing salt. Place the synthetic row immediately after its
-source user row. Entirely synthetic messages continue to reuse their original message ID.
-
 Use the V1 compaction user message ID as the ID of the collapsed V2 compaction message. This matches V2's use of the
 admitted compaction input ID and preserves references to the initiating message.
 
@@ -219,13 +64,9 @@ serialize the retained V1 tail beginning at `tail_start_id` for `recent`. Use an
 retained, and use the compaction user message creation time. Do not emit the paired summary assistant as a separate V2
 assistant row.
 
-Do not project incomplete or failed V1 compactions into `session_message`. Omit both the internal compaction user marker
-and its paired summary assistant when no successful summary was completed. Assign final sequence numbers after filtering
-so omitted compactions leave no gaps. Their source rows remain preserved in the V1 `message` and `part` tables.
-
-After rebuilding a session's `session_message`, replace its `event_sequence` watermark with that session's maximum
-backfilled `session_message.seq`. This prevents new V2 events from reusing sequence numbers or sorting before migrated
-history. The migrated session's prior `event` rows are removed in the same transaction.
+After rebuilding `session_message`, seed `event_sequence` with one row per migrated session. Set its watermark to that
+session's maximum backfilled `session_message.seq`. This prevents new V2 events from reusing sequence numbers or sorting
+before migrated history. The `event` table remains empty.
 
 ## Drop
 
@@ -233,7 +74,6 @@ Drop these pre-launch V2 tables without preserving or transforming their rows:
 
 - `session_input`
 - `session_context_epoch`
-- `data_migration`
 
 Do not transfer `session_input` rows into `session_pending`.
 
@@ -263,36 +103,16 @@ schema.
 New nullable session columns, including `fork_session_id`, `fork_boundary`, and `time_suspended`, require no explicit
 backfill. Existing rows naturally receive `NULL` when the generated migration adds the columns.
 
-## Execution
+## Verification
 
-Before transforming V1 rows, look for `opencode-next.db` in the data directory. This file was used by pre-launch V2
-builds. Open it read-only with Bun SQLite and copy its `project`, `session`, and `session_message` rows directly into the
-current `project`, `session_v2`, and `session_message` tables. Existing current projects and Sessions win ID collisions.
-Do not copy its durable events or runtime caches; initialize each imported Session's `event_sequence` watermark from its
-maximum message sequence. Commit each imported Session independently and leave the source database untouched.
+The canonical migration test should seed representative V1 sessions, messages, parts, todos, projects, accounts,
+credentials, permissions, shares, and workspaces. After migration, it should verify:
 
-The previous V2 import is part of this migration and uses the same completion marker. It needs no source-specific cursor:
-the destination Session row is the per-Session idempotency boundary, so a retry skips transactions that already committed.
-
-Store V1 backfill state in `kv`; do not retain a dedicated `data_migration` table. Store the last successfully migrated
-session ID under `migration.v1-v2.session.cursor` and write `migration.v1-v2.completed` with value `true` after every
-session finishes. Delete the cursor key on completion and return immediately on later calls when the completion key
-exists.
-
-Absence of the completion key means migration is required, including on a fresh database. Running the endpoint against a
-database with no sessions completes immediately and writes the completion key; fresh database initialization does not
-seed migration state specially.
-
-Process sessions in stable ID order. Rebuild one session in one transaction, including its `session_message` rows,
-session-level backfills, `event_sequence` watermark, and cursor update. If interrupted during a session, that transaction
-rolls back and the next endpoint call retries the same session. If it committed, the next call continues after the stored
-cursor. Mark the migration complete after the final session and return immediately on later calls.
-
-Ensure the global project exists using the current platform's filesystem root as its worktree. Process every `session`
-row, including archived, root, child, and empty sessions, as well as sessions whose messages are all skipped or internal.
-Reassign beta and V1 Sessions whose referenced project row is missing to the global project and log a warning. Each
-successfully committed session advances the cursor.
-
-## Testing
-
-Detailed migration test design is deferred until after the canonical migration is implemented.
+- Preserved rows and encoded values remain unchanged.
+- Todo rows remain available in the unchanged `todo` table.
+- `event` is empty, and stale pre-launch rows are absent from the rebuilt projections.
+- Backfilled `session_message` rows represent the canonical V1 `message` and `part` history.
+- Each migrated session's `event_sequence` watermark matches its maximum backfilled message sequence.
+- Dropped tables no longer exist.
+- New tables exist and are empty.
+- The final schema has no ungenerated changes.

infra/console.ts

@@ -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 }],
         },
       },
     },

install

@@ -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 ""

nix/hashes.json

@@ -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-RFek0QoEEjsgbqmTE/SxQAmPtYyzs0IPR2ugFn5Okrs=",
+    "aarch64-linux": "sha256-BmAxapY1YrAFn7mVq3/6A9+6Au5UIvSqBboHMkyJH3I=",
+    "aarch64-darwin": "sha256-Sx3bGWQqLlgoa/RudJxanjSzhFRNklckT2ffnO2I5F4=",
+    "x86_64-darwin": "sha256-CMOhiisHNowg06qadvgg4K+60zrynglwiT0qKYQ4NiA="
   }
 }

package.json

@@ -46,9 +46,9 @@
       "@octokit/rest": "22.0.0",
       "@hono/standard-validator": "0.2.0",
       "@hono/zod-validator": "0.4.2",
-      "@opentui/core": "0.5.2",
-      "@opentui/keymap": "0.5.2",
-      "@opentui/solid": "0.5.2",
+      "@opentui/core": "0.4.5",
+      "@opentui/keymap": "0.4.5",
+      "@opentui/solid": "0.4.5",
       "@tanstack/solid-virtual": "3.13.32",
       "@shikijs/stream": "4.2.0",
       "@standard-schema/spec": "1.1.0",
@@ -78,7 +78,7 @@
       "fuzzysort": "3.1.0",
       "get-east-asian-width": "1.6.0",
       "luxon": "3.6.1",
-      "marked": "18.0.7",
+      "marked": "17.0.6",
       "marked-shiki": "1.2.1",
       "remend": "1.3.0",
       "@playwright/test": "1.59.1",
@@ -100,7 +100,6 @@
       "@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"
     }
@@ -158,8 +157,6 @@
     "effect": "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",
     "@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",
@@ -171,8 +168,6 @@
     "@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",
-    "@cloudflare/vitest-pool-workers@0.12.6": "patches/@cloudflare%2Fvitest-pool-workers@0.12.6.patch",
-    "@ff-labs/fff-bun@0.10.1": "patches/@ff-labs%2Ffff-bun@0.10.1.patch"
+    "@tanstack/virtual-core@3.17.3": "patches/@tanstack%2Fvirtual-core@3.17.3.patch"
   }
 }

packages/ai/AGENTS.md

@@ -80,7 +80,7 @@ Route defaults are request-shaping defaults such as `headers`, `limits`, `genera
 
 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 ships a non-HTTP transport (OpenAI's WebSocket Responses backend, hypothetical bidirectional streaming APIs), the seam is `Transport` — `WebSocketTransport.jsonTransport.with(...)` constructs an IO template whose `prepare` receives the route endpoint/auth at compile time, builds a WebSocket URL and message, and whose `frames` yields decoded text from the socket. Same protocol and endpoint source, different transport.
+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
 
@@ -106,7 +106,7 @@ 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`, `responsesWebSocket`, and `chat`.
+- 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`.
@@ -124,11 +124,10 @@ import { model } from "@opencode-ai/ai/providers/openai/responses"
 
 const selected = model("gpt-5", {
   apiKey,
-  transport: "websocket",
 })
 ```
 
-Keep semantic APIs as separate entrypoints, such as OpenAI `chat` and `responses`. Keep transport choices inside the semantic entrypoint settings, so OpenAI Responses HTTP and WebSocket share one entrypoint. Provider facades may still expose named selectors such as `responsesWebSocket` for direct typed call sites; the package-like contract maps its settings to those selectors before returning an executable `LanguageModel`.
+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(...)`.
 
@@ -154,14 +153,16 @@ packages/ai/src/
     auth-options.ts         ProviderAuthOption shape, AuthOptions.bearer, AtLeastOne helper
     framing.ts              Framing type + Framing.sse
     transport/              transport implementations
-      index.ts              Transport type + HttpTransport / WebSocketTransport namespaces
+      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          WebSocketTransport.json + WebSocketExecutor service
+      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
-    openai-responses.ts       OpenAI tools/events/transports composed over OpenResponses
+    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

packages/ai/README.md

@@ -315,7 +315,6 @@ import { model } from "@opencode-ai/ai/providers/openai/responses"
 
 const selected = model("gpt-5", {
   apiKey: process.env.OPENAI_API_KEY,
-  transport: "websocket",
   headers: { "x-application": "opencode" },
   limits: { context: 200_000, output: 64_000 },
 })
@@ -332,7 +331,7 @@ OpenAI Chat and OpenAI Responses are separate semantic entrypoints:
 - `@opencode-ai/ai/providers/google-vertex/responses`
 - `@opencode-ai/ai/providers/google-vertex/messages`
 
-Responses HTTP versus WebSocket is a scoped `transport` setting on the OpenAI Responses entrypoint, not another entrypoint. 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, defaults, and transports. 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.
+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`.
 
@@ -368,12 +367,11 @@ Other provider exports listed above remain direct facades until they explicitly
 
 ## Provider options & HTTP overlays
 
-Request options in order of stability:
+Three escape hatches 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.
+2. **`providerOptions: { <provider>: {...} }`** — typed-at-the-facade provider-specific knobs (OpenAI `promptCacheKey`, Anthropic `thinking`, Gemini `thinkingConfig`, OpenRouter routing).
+3. **`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.
 

packages/ai/STATUS.md

@@ -1,6 +1,6 @@
 # LLM Provider Parity Status
 
-Last reviewed: 2026-07-24
+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.
 
@@ -16,8 +16,7 @@ This file tracks the gap between the native `@opencode-ai/ai` package and the AI
 | 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 HTTP              | `src/protocols/open-responses.ts`, `src/protocols/openai-responses.ts`, `src/providers/openai.ts`                                 | Usable. Extends the Open Responses baseline with hosted-tool event surfacing, reasoning replay metadata, GPT-5 defaults, and cache usage.                                | No explicit `previous_response_id` path. Typed options cover only a subset of Responses fields. Structured output is still mostly synthetic-tool based.                                                                                                   |
-| OpenAI Responses WebSocket         | `src/protocols/openai-responses.ts`, `src/route/transport/websocket.ts`                                                           | Present as `OpenAI.responsesWebSocket(...)`.                                                                                                                             | Runner/catalog support explicitly must not downgrade WebSocket routes; broader runtime selection is not complete.                                                                                                                                         |
+| 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.                                                                                                                                                                                                                  |
@@ -48,19 +47,19 @@ Other `aisdk:` packages, including Google Vertex, Azure, and Bedrock, currently
 
 ## AI SDK Package Parity Matrix
 
-| AI SDK package                    | Intended native target                                         | Status           | Biggest gaps                                                                                                                                                                                                        |
-| --------------------------------- | -------------------------------------------------------------- | ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| `@ai-sdk/openai`                  | `OpenAI.chat`, `OpenAI.responses`, `OpenAI.responsesWebSocket` | Partial / usable | Add complete typed option coverage, structured output strategy, explicit Responses continuation support, and runner route selection between Chat/Responses/WebSocket.                                               |
-| `@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.                                                                             |
+| 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
 
@@ -78,24 +77,24 @@ Other `aisdk:` packages, including Google Vertex, Azure, and Bedrock, currently
 
 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/WebSocket selected through settings. |
-| 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.                                    |
+| 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
 

packages/ai/example/call-sites.md

@@ -67,7 +67,6 @@ Examples:
 ```ts
 OpenAI.responses("gpt-4o")
 OpenAI.chat("gpt-4o")
-OpenAI.responsesWebSocket("gpt-4o")
 
 Azure.configure({ resourceName, apiKey }).responses("my-deployment")
 AmazonBedrock.configure({ region, credentials }).model("anthropic.claude-3-5-sonnet-20241022-v2:0")
@@ -250,11 +249,6 @@ const openAIChat = Route.make({
   auth: Auth.envBearer("OPENAI_API_KEY"),
 })
 
-const openAIResponsesWebSocket = openAIResponses.with({
-  id: "openai-responses-websocket",
-  transport: WebSocketTransport.json,
-})
-
 const openAIConfig = (input: OpenAIConfig) => ({
   endpoint: input.endpoint,
   auth: input.auth ?? (input.apiKey ? Auth.bearer(input.apiKey) : undefined),
@@ -266,13 +260,11 @@ const openAIConfig = (input: OpenAIConfig) => ({
 
 const configureOpenAI = (input: OpenAIConfig = {}) => {
   const responses = openAIResponses.with(openAIConfig(input))
-  const responsesWebSocket = openAIResponsesWebSocket.with(openAIConfig(input))
   const chat = openAIChat.with(openAIConfig(input))
 
   return {
     id: openAIProvider,
     responses: responses.model,
-    responsesWebSocket: responsesWebSocket.model,
     chat: chat.model,
     model: responses.model,
     configure: configureOpenAI,
@@ -342,22 +334,19 @@ const response =
   )
 ```
 
-For direct provider-facade calls, HTTP versus WebSocket is represented as named
-route selectors, not as model or request overrides. Same protocol, different
-transport, different route:
+For direct provider-facade calls, Responses has one semantic model and route:
 
 ```ts
 OpenAI.responses("gpt-4o")
-OpenAI.responsesWebSocket("gpt-4o")
 ```
 
-The package-like OpenAI Responses entrypoint instead keeps transport scoped to
-Responses settings while preserving the same `model(...)` contract:
+The package-like OpenAI Responses entrypoint has the same transport-neutral
+`model(...)` contract:
 
 ```ts
 import { model } from "@opencode-ai/ai/providers/openai/responses"
 
-model("gpt-4o", { apiKey, transport: "websocket" })
+model("gpt-4o", { apiKey })
 ```
 
 Vertex keeps Gemini, Chat, Responses, and Messages as separate package-like entrypoints,
@@ -387,11 +376,9 @@ import { model } from "@opencode-ai/ai/providers/google-vertex/messages"
 model("claude-sonnet-4-6", { project, location: "global" })
 ```
 
-The client should not require a different public layer just because a selected
-route uses WebSocket. Use one `LLMClient.layer` with HTTP and WebSocket runtime
-capabilities available; routes that do not need WebSocket simply never touch it.
-If a WebSocket route is selected in an environment without WebSocket support,
-fail with a typed transport configuration error.
+The client does not require a different public layer for WebSocket execution.
+Responses routes use HTTP by default, and callers may pass a channel executor per
+call. Routes without channel support simply ignore that execution capability.
 
 Azure is a route specialization with auth/path/default changes plus input
 mapping. The public API configures the Azure resource once, then selects
@@ -499,16 +486,13 @@ generic dynamic resolver:
 const model =
   providerID === "azure"
     ? Azure.configure(resolvedAzureConfig).responses(apiModelID)
-    : endpoint.websocket
-      ? OpenAI.responsesWebSocket(apiModelID)
-      : OpenAI.responses(apiModelID)
+    : OpenAI.responses(apiModelID)
 ```
 
 That boundary can branch on durable config/catalog metadata and call typed
-provider APIs directly. A direct provider-facade boundary maps metadata like
-`endpoint.websocket` to `OpenAI.responsesWebSocket(apiModelID)`. A package-loading
-boundary passes `transport: "websocket"` to the OpenAI Responses entrypoint.
-The client runtime only executes the route carried by the resulting model.
+provider APIs directly. Transport selection remains execution policy: a Session
+or other caller may pass a WebSocket channel executor per call without changing
+the model constructed by this boundary.
 
 ## Competitive Shape
 
@@ -544,9 +528,8 @@ App boundary = explicit durable-config -> typed-provider call
   id.
 - No `model(id, overrides)` escape hatch. Model selection takes the model id;
   endpoint/auth/deployment customization happens by configuring the route first.
-- No transport override on an executable model or request. Direct provider
-  facades use `responses` versus `responsesWebSocket`; the package-like Responses
-  entrypoint maps its scoped `transport` setting before constructing the model.
+- No transport setting on a provider or executable model. OpenAI Responses uses
+  HTTP by default and accepts an optional per-call channel executor as execution policy.
 - No separate public `LLMClient.layerWithWebSocket`. The runtime should expose one
   client layer with the available transport capabilities.
 - No executable `ModelRef`. The executable handle is `LanguageModel`; durable model
@@ -580,12 +563,10 @@ App boundary = explicit durable-config -> typed-provider call
 - [x] Make unconfigured transports reusable constants such as
       `HttpTransport.sseJson`; keep transport functions only for configured/fresh
       state construction.
-- [x] Collapse the public WebSocket runtime split so one `LLMClient.layer`
-      exposes available transport capabilities and selected routes fail with typed
-      transport config errors when a required capability is missing.
+- [x] Collapse the public WebSocket runtime split so one `LLMClient.layer` accepts
+      optional per-call channel execution without changing route identity.
 - [x] Convert OpenAI provider APIs to provider-facade shape:
-      `OpenAI.configure(config).responses(id)`, `.chat(id)`, and
-      `.responsesWebSocket(id)`.
+      `OpenAI.configure(config).responses(id)` and `.chat(id)`.
 - [x] Convert Azure to a configured facade where resource/base URL/api version
       setup happens before selecting deployment ids.
 - [x] Split Cloudflare products into separate facades such as
@@ -599,10 +580,8 @@ App boundary = explicit durable-config -> typed-provider call
 - [ ] Decide whether a tiny `Provider.define(...)` helper is warranted after two
       or three provider conversions; start with plain objects if duplication is not
       yet painful.
-- [x] Update `packages/opencode/src/session/llm/native-request.ts` to construct
-      executable models at the session boundary with explicit provider facade
-      calls, mapping catalog metadata such as `endpoint.websocket` to the correct
-      named route selector.
+- [x] Keep executable model construction transport-neutral at the Session boundary;
+      Session-scoped execution policy supplies channel capability separately.
 - [ ] Update tests so direct route/provider tests assert route values are carried
       by executable models, and opencode/native tests assert boundary-based route
       selection.

packages/ai/example/tutorial.ts

@@ -1,6 +1,6 @@
 import { Config, Effect, Formatter, Layer, Schema, Stream } from "effect"
 import { LLM, LLMClient, LLMRequest, Message, ProviderID, Tool, ToolRuntime } from "@opencode-ai/ai"
-import { Route, Auth, Endpoint, Framing, Protocol, RequestExecutor, WebSocketExecutor } from "@opencode-ai/ai/route"
+import { Route, Auth, Endpoint, Framing, Protocol, RequestExecutor } from "@opencode-ai/ai/route"
 import { OpenAI } from "@opencode-ai/ai/providers"
 
 /**
@@ -33,10 +33,9 @@ const model = OpenAI.configure({
 //
 //   - `generation`: common controls such as max tokens, temperature, topP/topK,
 //     penalties, seed, and stop sequences.
-//   - `promptCacheKey`: stable cache affinity for protocols that support it.
 //   - `providerOptions`: namespaced provider-native behavior. For example,
-//     OpenAI store behavior, Anthropic thinking, Gemini thinking config, or
-//     OpenRouter routing/reasoning.
+//     OpenAI cache keys and store behavior, Anthropic thinking, Gemini thinking
+//     config, or OpenRouter routing/reasoning.
 //   - `http`: last-resort serializable overlays for final request body, headers,
 //     and query params. Prefer typed `providerOptions` when a field is stable.
 //
@@ -46,7 +45,9 @@ const request = LLM.request({
   system: "You are concise and practical.",
   prompt: "Tell me a joke",
   generation: { maxTokens: 80, temperature: 0.7 },
-  promptCacheKey: "tutorial-joke",
+  providerOptions: {
+    openai: { promptCacheKey: "tutorial-joke" },
+  },
 })
 
 // 3. `generate` sends the request and collects the event stream into one
@@ -213,8 +214,7 @@ const FakeEcho = {
 // enabled at a time so the tutorial can demonstrate generate, stream, or
 // tool-loop behavior without spending tokens on every example.
 const requestExecutorLayer = RequestExecutor.fetchLayer
-const llmDeps = Layer.mergeAll(requestExecutorLayer, WebSocketExecutor.layer)
-const llmClientLayer = LLMClient.layer.pipe(Layer.provide(llmDeps))
+const llmClientLayer = LLMClient.layer.pipe(Layer.provide(requestExecutorLayer))
 
 const program = Effect.gen(function* () {
   // yield* generateOnce
@@ -222,6 +222,6 @@ const program = Effect.gen(function* () {
   // yield* generateStructuredObject
   // yield* generateDynamicObject.pipe(Effect.andThen((response) => Effect.sync(() => console.log(response.object))))
   yield* streamWithTools
-}).pipe(Effect.provide(Layer.mergeAll(llmDeps, llmClientLayer)))
+}).pipe(Effect.provide(Layer.mergeAll(requestExecutorLayer, llmClientLayer)))
 
 Effect.runPromise(program)

packages/ai/script/recording-cost-report.ts

@@ -2,7 +2,7 @@ import * as fs from "node:fs/promises"
 import * as path from "node:path"
 
 const RECORDINGS_DIR = path.resolve(import.meta.dir, "..", "test", "fixtures", "recordings")
-const MODELS_DEV_URL = "https://models.opencode.ai/api.json"
+const MODELS_DEV_URL = "https://models.dev/api.json"
 
 type JsonRecord = Record<string, unknown>
 

packages/ai/src/cache-policy.ts

@@ -10,8 +10,8 @@
 //
 // 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"
+import { CacheHint, type CachePolicy, type CachePolicyObject } from "./schema/options"
+import { LLMRequest, Message, ToolDefinition, type ContentPart } from "./schema/messages"
 
 const AUTO: CachePolicyObject = {
   tools: true,
@@ -133,8 +133,7 @@ const countHints = (request: LLMRequest) =>
 
 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
+  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
 

packages/ai/src/image-client.ts

@@ -1,7 +1,7 @@
 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"
+import { RequestExecutor } from "./route/executor"
+import type { ImageOptions, ImageRequest, ImageRequestFor, ImageResponse } from "./image"
+import type { AIError } from "./schema"
 
 export type Execute = RequestExecutor.Interface["execute"]
 

packages/ai/src/image.ts

@@ -1,14 +1,6 @@
 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"
+import { HttpOptions, InvalidRequestReason, AIError, ModelID, ProviderID, ProviderMetadata, Usage } from "./schema"
+import { ImageClient, Service, type Execute as ImageExecute } from "./image-client"
 
 export interface ImageRoute<Options extends ImageOptions = ImageOptions> {
   readonly id: string

packages/ai/src/index.ts

@@ -1,22 +1,22 @@
-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 { LLMClient } from "./route/client"
+export { ImageClient } from "./image-client"
+export { Auth } from "./route/auth"
+export { Provider } from "./provider"
+export { ProviderPackage } from "./provider-package"
+export { isContextOverflow, isContextOverflowFailure } from "./provider-error"
 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"
+} from "./route/client"
+export * from "./schema"
+export { GeneratedImage, ImageInput, ImageInputSchema, ImageModel, ImageRequest, ImageResponse } from "./image"
+export type { ImageModelOptions, ImageOptions, ImageRequestFor, ImageRequestInput, ImageRoute } from "./image"
+export { Image } from "./image"
+export { Tool, ToolFailure, toDefinitions } from "./tool"
+export { ToolRuntime } from "./tool-runtime"
+export type { DispatchResult as ToolDispatchResult, ToolSettlement } from "./tool-runtime"
 export type {
   AnyExecutableTool,
   AnyTool,
@@ -29,14 +29,11 @@ export type {
   Tools,
   ToolSchema,
   ToolToModelOutput,
-} from "./tool.js"
-export * as LLM from "./llm.js"
+} from "./tool"
+export * as LLM from "./llm"
 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"
+} from "./provider"
+export type { Definition as ProviderPackageDefinition, Settings as ProviderPackageSettings } from "./provider-package"

packages/ai/src/llm.ts

@@ -1,5 +1,5 @@
 import { Effect, JsonSchema, Schema } from "effect"
-import { LLMClient, Service } from "./route/client.js"
+import { LLMClient, Service } from "./route/client"
 import {
   GenerationOptions,
   HttpOptions,
@@ -15,8 +15,8 @@ import {
   ToolDefinition,
   type ContentPart,
   type LanguageModelProviderOptions,
-} from "./schema/index.js"
-import { make as makeTool, toDefinitions, type ToolSchema } from "./tool.js"
+} from "./schema"
+import { make as makeTool, toDefinitions, type ToolSchema } from "./tool"
 
 /** Input accepted by `LLM.request`, normalized into the canonical `LLMRequest` class. */
 export type RequestInput<SelectedLanguageModel extends LanguageModel = LanguageModel> = Omit<

packages/ai/src/protocols.ts

@@ -1 +1 @@
-export * from "./protocols/index.js"
+export * from "./protocols/index"

packages/ai/src/protocols/anthropic-messages.ts

@@ -1,10 +1,10 @@
 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 { Route } from "../route/client"
+import { Auth } from "../route/auth"
+import { Endpoint } from "../route/endpoint"
+import { Framing } from "../route/framing"
+import { Protocol } from "../route/protocol"
 import {
   AIError,
   LLMEvent,
@@ -21,13 +21,13 @@ import {
   type ToolCallPart,
   type ToolDefinition,
   type ToolResultPart,
-} from "../schema/index.js"
-import { JsonObject, optionalArray, optionalNull, ProviderShared } from "./shared.js"
-import { classifyProviderFailure } from "../provider-error.js"
-import * as Cache from "./utils/cache.js"
-import { Lifecycle } from "./utils/lifecycle.js"
-import { ToolSchemaProjection } from "./utils/tool-schema.js"
-import { ToolStream } from "./utils/tool-stream.js"
+} from "../schema"
+import { JsonObject, optionalArray, optionalNull, ProviderShared } from "./shared"
+import { classifyProviderFailure } from "../provider-error"
+import * as Cache from "./utils/cache"
+import { Lifecycle } from "./utils/lifecycle"
+import { ToolSchemaProjection } from "./utils/tool-schema"
+import { ToolStream } from "./utils/tool-stream"
 
 const ADAPTER = "anthropic-messages"
 const MEDIA_MIMES = new Set<string>([...ProviderShared.IMAGE_MIMES, ...ProviderShared.PDF_MIMES])
@@ -422,12 +422,14 @@ const lowerMedia = Effect.fn("AnthropicMessages.lowerMedia")(function* (part: Me
 
 // Tool results may carry structured text, images, and documents. Keep media as provider-native
 // content instead of JSON-stringifying base64 into a prompt string.
-const lowerToolResultContentItem = Effect.fnUntraced(function* (item: Tool.Content) {
+const lowerToolResultContentItem = Effect.fn("AnthropicMessages.lowerToolResultContentItem")(function* (
+  item: Tool.Content,
+) {
   if (item.type === "text") return { type: "text" as const, text: item.text } satisfies AnthropicTextBlock
   return yield* lowerMedia({ type: "media", mediaType: item.mime, data: item.uri, filename: item.name })
 })
 
-const lowerToolResultContent = Effect.fnUntraced(function* (part: ToolResultPart) {
+const lowerToolResultContent = Effect.fn("AnthropicMessages.lowerToolResultContent")(function* (part: ToolResultPart) {
   // Text / json / error results stay as a string for backward compatibility
   // with existing cassettes and provider expectations.
   if (part.result.type !== "content") return ProviderShared.toolResultText(part)
@@ -573,10 +575,7 @@ const lowerMessages = Effect.fn("AnthropicMessages.lowerMessages")(function* (
         cache_control: cacheControl(breakpoints, part.cache),
       })
     }
-    const previous = messages.at(-1)
-    if (previous?.role === "user" && previous.content.every((block) => block.type === "tool_result"))
-      messages[messages.length - 1] = { role: "user", content: [...previous.content, ...content] }
-    else messages.push({ role: "user", content })
+    messages.push({ role: "user", content })
   }
 
   return messages
@@ -1032,4 +1031,4 @@ export const route = Route.make({
   headers: () => ({ "anthropic-version": "2023-06-01" }),
 })
 
-export * as AnthropicMessages from "./anthropic-messages.js"
+export * as AnthropicMessages from "./anthropic-messages"

packages/ai/src/protocols/bedrock-converse.ts

@@ -1,7 +1,7 @@
 import { Effect, Schema } from "effect"
-import { Route } from "../route/client.js"
-import { Endpoint } from "../route/endpoint.js"
-import { Protocol } from "../route/protocol.js"
+import { Route } from "../route/client"
+import { Endpoint } from "../route/endpoint"
+import { Protocol } from "../route/protocol"
 import {
   AIError,
   LLMEvent,
@@ -17,20 +17,20 @@ import {
   type ToolCallPart,
   type ToolDefinition,
   type ToolResultPart,
-} from "../schema/index.js"
-import { BedrockEventStream } from "./bedrock-event-stream.js"
-import { classifyProviderFailure } from "../provider-error.js"
-import { JsonObject, optionalArray, ProviderShared } from "./shared.js"
-import { BedrockAuth } from "./utils/bedrock-auth.js"
-import { BedrockCache } from "./utils/bedrock-cache.js"
-import { BedrockMedia } from "./utils/bedrock-media.js"
-import { Lifecycle } from "./utils/lifecycle.js"
-import { ToolSchemaProjection } from "./utils/tool-schema.js"
-import { ToolStream } from "./utils/tool-stream.js"
+} from "../schema"
+import { BedrockEventStream } from "./bedrock-event-stream"
+import { classifyProviderFailure } from "../provider-error"
+import { JsonObject, optionalArray, ProviderShared } from "./shared"
+import { BedrockAuth } from "./utils/bedrock-auth"
+import { BedrockCache } from "./utils/bedrock-cache"
+import { BedrockMedia } from "./utils/bedrock-media"
+import { Lifecycle } from "./utils/lifecycle"
+import { ToolSchemaProjection } from "./utils/tool-schema"
+import { ToolStream } from "./utils/tool-stream"
 
 const ADAPTER = "bedrock-converse"
 
-export type { Credentials as BedrockCredentials } from "./utils/bedrock-auth.js"
+export type { Credentials as BedrockCredentials } from "./utils/bedrock-auth"
 
 // =============================================================================
 // Request Body Schema
@@ -348,10 +348,7 @@ const lowerMessages = Effect.fn("BedrockConverse.lowerMessages")(function* (
           continue
         }
       }
-      const previous = messages.at(-1)
-      if (previous?.role === "user")
-        messages[messages.length - 1] = { role: "user", content: [...previous.content, ...content] }
-      else messages.push({ role: "user", content })
+      messages.push({ role: "user", content })
       continue
     }
 
@@ -395,10 +392,7 @@ const lowerMessages = Effect.fn("BedrockConverse.lowerMessages")(function* (
       const cachePoint = BedrockCache.block(breakpoints, part.cache)
       if (cachePoint) content.push(cachePoint)
     }
-    const previous = messages.at(-1)
-    if (previous?.role === "user")
-      messages[messages.length - 1] = { role: "user", content: [...previous.content, ...content] }
-    else messages.push({ role: "user", content })
+    messages.push({ role: "user", content })
   }
 
   return messages
@@ -737,4 +731,4 @@ export const route = Route.make({
 
 export const sigV4Auth = BedrockAuth.sigV4
 
-export * as BedrockConverse from "./bedrock-converse.js"
+export * as BedrockConverse from "./bedrock-converse"

packages/ai/src/protocols/bedrock-event-stream.ts

@@ -1,8 +1,8 @@
 import { EventStreamCodec } from "@smithy/eventstream-codec"
 import { fromUtf8, toUtf8 } from "@smithy/util-utf8"
 import { Effect, Stream } from "effect"
-import { Framing } from "../route/framing.js"
-import { ProviderShared } from "./shared.js"
+import { Framing } from "../route/framing"
+import { ProviderShared } from "./shared"
 
 // Bedrock streams responses using the AWS event stream binary protocol — each
 // frame is `[length:4][headers-length:4][prelude-crc:4][headers][payload][crc:4]`.
@@ -98,4 +98,4 @@ export const framing = (route: string): Framing.Definition<object> => ({
   frame: (bytes) => bytes.pipe(Stream.mapAccumEffect(() => initialFrameBuffer, consumeFrames(route))),
 })
 
-export * as BedrockEventStream from "./bedrock-event-stream.js"
+export * as BedrockEventStream from "./bedrock-event-stream"

packages/ai/src/protocols/gemini.ts

@@ -1,10 +1,10 @@
 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 { Route } from "../route/client"
+import { Auth } from "../route/auth"
+import { Endpoint } from "../route/endpoint"
+import { Framing } from "../route/framing"
+import { Protocol } from "../route/protocol"
 import {
   LLMEvent,
   Usage,
@@ -17,28 +17,16 @@ import {
   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"
+} from "../schema"
+import { JsonObject, optionalArray, ProviderShared } from "./shared"
+import { GeminiToolSchema } from "./utils/gemini-tool-schema"
+import { Lifecycle } from "./utils/lifecycle"
+import { ToolSchemaProjection } from "./utils/tool-schema"
 
 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(?:[.-]|$)/i.test(modelID)
-}
-
 export interface OptionsInput {
   readonly [key: string]: unknown
   readonly cachedContent?: string
@@ -157,9 +145,6 @@ const GeminiGenerationConfig = Schema.Struct({
   temperature: Schema.optional(Schema.Number),
   topP: Schema.optional(Schema.Number),
   topK: Schema.optional(Schema.Number),
-  frequencyPenalty: Schema.optional(Schema.Number),
-  presencePenalty: Schema.optional(Schema.Number),
-  seed: Schema.optional(Schema.Number),
   stopSequences: optionalArray(Schema.String),
   thinkingConfig: Schema.optional(GeminiThinkingConfig),
 })
@@ -217,13 +202,11 @@ interface ParserState {
 //    keys on non-object scalars. Mirrors OpenCode's historical Gemini rules.
 //
 // 2. Project — lossy mapping from JSON Schema to Gemini's schema dialect:
-//    drop empty root parameter schemas while preserving nested empty objects,
-//    expand type arrays into `anyOf`, derive `nullable: true` from null members,
-//    coerce `const` to `[const]` enum, recurse properties/items, and propagate
+//    drop empty objects, derive `nullable: true` from `type: [..., "null"]`,
+//    coerce `const` to `[const]` enum, recurse properties/items, propagate
 //    only an allowlisted set of keys (description, required, format, type,
-//    nullable, enum, properties, items, allOf, anyOf, oneOf, minLength).
-//    Anything outside the allowlist (e.g. `additionalProperties`, `$ref`) is
-//    silently dropped.
+//    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
@@ -299,8 +282,6 @@ const lowerMessages = Effect.fn("Gemini.lowerMessages")(function* (request: LLMR
 
     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"])
@@ -313,17 +294,7 @@ const lowerMessages = Effect.fn("Gemini.lowerMessages")(function* (request: LLMR
           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
+          parts.push(lowerToolCall(part))
           continue
         }
       }
@@ -417,9 +388,6 @@ const fromRequest = Effect.fn("Gemini.fromRequest")(function* (request: LLMReque
     temperature: generation?.temperature,
     topP: generation?.topP,
     topK: generation?.topK,
-    frequencyPenalty: generation?.frequencyPenalty,
-    presencePenalty: generation?.presencePenalty,
-    seed: generation?.seed,
     stopSequences: generation?.stop,
     thinkingConfig: options.thinkingConfig,
   }
@@ -643,4 +611,4 @@ export const route = Route.make({
   framing: Framing.sse,
 })
 
-export * as Gemini from "./gemini.js"
+export * as Gemini from "./gemini"

packages/ai/src/protocols/google-images.ts

@@ -7,8 +7,8 @@ import {
   type ImageInput,
   type ImageRequestFor,
   type ImageRoute,
-} from "../image.js"
-import { Auth, type Definition as AuthDefinition } from "../route/auth.js"
+} from "../image"
+import { Auth, type Definition as AuthDefinition } from "../route/auth"
 import {
   InvalidProviderOutputReason,
   AIError,
@@ -17,9 +17,9 @@ import {
   mergeJsonRecords,
   type HttpOptions,
   type ProviderMetadata,
-} from "../schema/index.js"
-import { ProviderShared } from "./shared.js"
-import { ImageInputs } from "./utils/image-input.js"
+} from "../schema"
+import { ProviderShared } from "./shared"
+import { ImageInputs } from "./utils/image-input"
 
 const ADAPTER = "google-images"
 export const DEFAULT_BASE_URL = "https://generativelanguage.googleapis.com/v1beta"

packages/ai/src/protocols/index.ts

@@ -1,9 +1,10 @@
-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 AnthropicMessages from "./anthropic-messages"
+export * as BedrockConverse from "./bedrock-converse"
+export * as Gemini from "./gemini"
+export * as OpenAIChat from "./openai-chat"
+export * as OpenAIImages from "./openai-images"
+export * as OpenAICompatibleChat from "./openai-compatible-chat"
+export * as OpenAICompatibleResponses from "./openai-compatible-responses"
+export * as OpenAIResponses from "./openai-responses"
+export * as OpenResponses from "./open-responses"
+export * as OpenResponsesChannel from "./open-responses-channel"

packages/ai/src/protocols/open-responses-channel.ts

@@ -0,0 +1,119 @@
+import { Effect, Schema, Stream } from "effect"
+import { Headers } from "effect/unstable/http"
+import { Framing } from "../route/framing"
+import {
+  HttpTransport,
+  WebSocketTransport,
+  type Transport,
+  type WebSocketChannelDriver,
+  type WebSocketChannelExchange,
+} from "../route/transport"
+import * as ProviderShared from "./shared"
+import { OpenResponses } from "./open-responses"
+
+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
+}
+
+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
+    return encodeMessage(yield* decodeMessage({ ...request, type: "response.create" }))
+  })
+
+const driver = (options: Options, body: string): WebSocketChannelDriver => ({
+  create: () => Effect.succeed({ message: body, mode: "full" }),
+  observe: (_create, frame) =>
+    Effect.gen(function* () {
+      const event = yield* decodeEvent(frame).pipe(
+        Effect.mapError(() => ProviderShared.eventError(options.id, `Invalid ${options.name} WebSocket event`, frame)),
+      )
+      if (event.type === "response.completed") return { type: "completed", frame }
+      if (event.type === "response.incomplete") return { type: "incomplete", frame }
+      if (event.type === "response.failed")
+        return {
+          type: "provider-failure",
+          error: OpenResponses.providerFailure(options.id, event, `${options.name} response failed`),
+        }
+      if (event.type === "error") {
+        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`),
+        }
+      }
+      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")
+        return {
+          http: {
+            request: ProviderShared.jsonPost({ url: parts.url, body: parts.bodyText, headers: parts.headers }),
+            framing: Framing.sse,
+            middleware: input.middleware,
+          },
+          channel: input.webSocket
+            ? {
+                url: yield* WebSocketTransport.toWebSocketUrl(parts.url),
+                headers,
+                rotateAfterMs: options.rotateAfterMs,
+                driver: driver(options, yield* message(parts.jsonBody)),
+              }
+            : undefined,
+        }
+      }),
+    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

packages/ai/src/protocols/open-responses.ts

@@ -1,7 +1,7 @@
 import { Effect, Schema } from "effect"
 import type { Content } from "@opencode-ai/schema/tool"
-import { HttpTransport } from "../route/transport/index.js"
-import { Protocol } from "../route/protocol.js"
+import { HttpTransport } from "../route/transport"
+import { Protocol } from "../route/protocol"
 import {
   AIError,
   LLMEvent,
@@ -16,13 +16,13 @@ import {
   type ToolCallPart,
   type ToolDefinition,
   type ToolResultPart,
-} from "../schema/index.js"
-import { JsonObject, optionalArray, optionalNull, ProviderShared } from "./shared.js"
-import { classifyProviderFailure } from "../provider-error.js"
-import { OpenResponsesOptions } from "./utils/open-responses-options.js"
-import { Lifecycle } from "./utils/lifecycle.js"
-import { ToolSchemaProjection } from "./utils/tool-schema.js"
-import { ToolStream } from "./utils/tool-stream.js"
+} from "../schema"
+import { JsonObject, optionalArray, optionalNull, ProviderShared } from "./shared"
+import { classifyProviderFailure } from "../provider-error"
+import { OpenResponsesOptions } from "./utils/open-responses-options"
+import { Lifecycle } from "./utils/lifecycle"
+import { ToolSchemaProjection } from "./utils/tool-schema"
+import { ToolStream } from "./utils/tool-stream"
 
 const ADAPTER = "open-responses"
 const NAME = "Open Responses"
@@ -211,11 +211,43 @@ export type StreamItem = Schema.Schema.Type<typeof StreamItem>
 // event-level `error` envelope, so accept all three shapes here.
 // https://www.openresponses.org/specification
 const OpenResponsesErrorPayload = Schema.Struct({
+  type: optionalNull(Schema.String),
   code: optionalNull(Schema.String),
   message: optionalNull(Schema.String),
   param: optionalNull(Schema.String),
 })
 
+const WebSocketErrorHeader = Schema.Union([Schema.String, Schema.Number, Schema.Boolean])
+export const WebSocketErrorEvent = Schema.StructWithRest(
+  Schema.Struct({
+    type: Schema.tag("error"),
+    status: Schema.optional(Schema.Number),
+    status_code: Schema.optional(Schema.Number),
+    code: optionalNull(Schema.String),
+    message: Schema.optional(Schema.String),
+    param: optionalNull(Schema.String),
+    error: optionalNull(OpenResponsesErrorPayload),
+    headers: Schema.optional(Schema.Record(Schema.String, WebSocketErrorHeader)),
+  }),
+  [Schema.Record(Schema.String, Schema.Unknown)],
+)
+const decodeWebSocketErrorEvent = Schema.decodeUnknownEffect(WebSocketErrorEvent)
+
+export const decodeKnownErrorEvent = (event: Event) =>
+  decodeWebSocketErrorEvent({
+    ...event,
+    status: typeof event.status === "number" ? event.status : undefined,
+    status_code: typeof event.status_code === "number" ? event.status_code : undefined,
+    headers: ProviderShared.isRecord(event.headers)
+      ? Object.fromEntries(
+          Object.entries(event.headers).filter(
+            (entry): entry is [string, string | number | boolean] =>
+              typeof entry[1] === "string" || typeof entry[1] === "number" || typeof entry[1] === "boolean",
+          ),
+        )
+      : undefined,
+  })
+
 export const Event = Schema.StructWithRest(
   Schema.Struct({
     type: Schema.String,
@@ -240,6 +272,9 @@ export const Event = Schema.StructWithRest(
     message: Schema.optional(Schema.String),
     param: optionalNull(Schema.String),
     error: optionalNull(OpenResponsesErrorPayload),
+    status: Schema.optional(Schema.Unknown),
+    status_code: Schema.optional(Schema.Unknown),
+    headers: Schema.optional(Schema.Unknown),
   }),
   [Schema.Record(Schema.String, Schema.Unknown)],
 )
@@ -358,7 +393,7 @@ const lowerMedia = Effect.fn("OpenResponses.lowerMedia")(function* (
   return { type: "input_image" as const, image_url: media.dataUrl }
 })
 
-const lowerUserContent = Effect.fnUntraced(function* (
+const lowerUserContent = Effect.fn("OpenResponses.lowerUserContent")(function* (
   part: LLMRequest["messages"][number]["content"][number],
   request: LLMRequest,
   extension: Extension,
@@ -370,7 +405,7 @@ const lowerUserContent = Effect.fnUntraced(function* (
 
 // Tool results may carry structured text, images, and files. Keep media as provider-native
 // content instead of JSON-stringifying base64 into a prompt string.
-const lowerToolResultContentItem = Effect.fnUntraced(function* (
+const lowerToolResultContentItem = Effect.fn("OpenResponses.lowerToolResultContentItem")(function* (
   item: Content,
   request: LLMRequest,
   extension: Extension,
@@ -383,7 +418,7 @@ const lowerToolResultContentItem = Effect.fnUntraced(function* (
   )
 })
 
-const lowerToolResultOutput = Effect.fnUntraced(function* (
+const lowerToolResultOutput = Effect.fn("OpenResponses.lowerToolResultOutput")(function* (
   part: ToolResultPart,
   request: LLMRequest,
   extension: Extension,
@@ -539,7 +574,7 @@ const lowerOptions = (request: LLMRequest) => {
   return {
     ...(options.instructions ? { instructions: options.instructions } : {}),
     ...(options.store !== undefined ? { store: options.store } : {}),
-    ...(request.promptCacheKey ? { prompt_cache_key: request.promptCacheKey } : {}),
+    ...(options.promptCacheKey ? { prompt_cache_key: options.promptCacheKey } : {}),
     ...(options.include ? { include: options.include } : {}),
     ...(options.reasoningEffort || options.reasoningSummary
       ? { reasoning: { effort: options.reasoningEffort, summary: options.reasoningSummary } }
@@ -632,9 +667,9 @@ export type StepResult = readonly [ParserState, ReadonlyArray<LLMEvent>]
 const NO_EVENTS: StepResult["1"] = []
 
 // `response.completed` / `response.incomplete` are clean finishes that emit a
-// `finish` event; `response.failed` is a hard failure. All three end the stream,
-// so keep this set aligned with `step` and the protocol's terminal predicate.
-const TERMINAL_TYPES = new Set(["response.completed", "response.incomplete", "response.failed"])
+// `finish` event; `response.failed` and `error` are hard failures. All four end
+// the stream, so keep this set aligned with `step` and the protocol's terminal predicate.
+const TERMINAL_TYPES = new Set(["error", "response.completed", "response.incomplete", "response.failed"])
 export const terminal = (event: Event) => TERMINAL_TYPES.has(event.type)
 
 const onOutputTextDelta = (state: ParserState, event: Event, id: string): StepResult => {
@@ -966,16 +1001,24 @@ const providerErrorMessage = (event: Event, fallback: string): string => {
   return message || code || fallback
 }
 
-const providerError = (state: ParserState, event: Event, fallback: string) => {
+export const providerFailure = (id: string, event: Event, fallback: string) => {
   const code = event.code || event.error?.code || event.response?.error?.code || undefined
   const message = providerErrorMessage(event, fallback)
+  const status =
+    typeof event.status === "number"
+      ? event.status
+      : typeof event.status_code === "number"
+        ? event.status_code
+        : undefined
   return new AIError({
-    module: state.id,
+    module: id,
     method: "stream",
-    reason: classifyProviderFailure({ message, code }),
+    reason: classifyProviderFailure({ message, code, status }),
   })
 }
 
+const providerError = (state: ParserState, event: Event, fallback: string) => providerFailure(state.id, event, fallback)
+
 export const step = (state: ParserState, event: Event) => {
   if (event.type === "response.output_text.delta" || event.type === "response.output_text.done") {
     if (!event.item_id) return ProviderShared.eventError(state.id, `${event.type} is missing item_id`)
@@ -1015,7 +1058,11 @@ export const step = (state: ParserState, event: Event) => {
   if (event.type === "response.completed" || event.type === "response.incomplete")
     return Effect.succeed(onResponseFinish(state, event))
   if (event.type === "response.failed") return providerError(state, event, `${state.name} response failed`)
-  if (event.type === "error") return providerError(state, event, `${state.name} stream error`)
+  if (event.type === "error")
+    return decodeKnownErrorEvent(event).pipe(
+      Effect.mapError(() => ProviderShared.eventError(state.id, `${state.name} returned a malformed error event`)),
+      Effect.flatMap(() => providerError(state, event, `${state.name} stream error`)),
+    )
   return Effect.succeed<StepResult>([state, NO_EVENTS])
 }
 
@@ -1061,4 +1108,4 @@ export const protocol = Protocol.make({
 
 export const httpTransport = HttpTransport.sseJson.with<OpenResponsesBody>()
 
-export * as OpenResponses from "./open-responses.js"
+export * as OpenResponses from "./open-responses"

packages/ai/src/protocols/openai-chat.ts

@@ -1,10 +1,10 @@
 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 { Route } from "../route/client"
+import { Auth } from "../route/auth"
+import { Endpoint } from "../route/endpoint"
+import { HttpTransport } from "../route/transport"
+import { Protocol } from "../route/protocol"
 import {
   AIError,
   LLMEvent,
@@ -19,13 +19,13 @@ import {
   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"
+} from "../schema"
+import { classifyProviderFailure } from "../provider-error"
+import { isRecord, JsonObject, optionalArray, optionalNull, ProviderShared } from "./shared"
+import { OpenAIOptions } from "./utils/openai-options"
+import { Lifecycle } from "./utils/lifecycle"
+import { ToolSchemaProjection } from "./utils/tool-schema"
+import { ToolStream } from "./utils/tool-stream"
 
 const ADAPTER = "openai-chat"
 const IMAGE_MIMES = new Set<string>(ProviderShared.IMAGE_MIMES)
@@ -67,12 +67,6 @@ const OpenAIChatAssistantToolCall = Schema.Struct({
 })
 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"),
@@ -132,7 +126,6 @@ export const bodyFields = {
   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),
@@ -152,33 +145,22 @@ export type OpenAIChatBody = Schema.Schema.Type<typeof OpenAIChatBody>
 // 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 OpenAIChatUsage = Schema.Struct({
+  prompt_tokens: Schema.optional(Schema.Number),
+  completion_tokens: Schema.optional(Schema.Number),
+  total_tokens: Schema.optional(Schema.Number),
+  prompt_tokens_details: optionalNull(
+    Schema.Struct({
+      cached_tokens: Schema.optional(Schema.Number),
+      cache_write_tokens: Schema.optional(Schema.Number),
+    }),
+  ),
+  completion_tokens_details: optionalNull(
+    Schema.Struct({
+      reasoning_tokens: Schema.optional(Schema.Number),
+    }),
+  ),
+})
 
 const OpenAIChatToolCallDeltaFunction = Schema.Struct({
   name: optionalNull(Schema.String),
@@ -186,7 +168,7 @@ const OpenAIChatToolCallDeltaFunction = Schema.Struct({
 })
 
 const OpenAIChatToolCallDelta = Schema.Struct({
-  index: optionalNull(Schema.Number),
+  index: Schema.Number,
   id: optionalNull(Schema.String),
   function: optionalNull(OpenAIChatToolCallDeltaFunction),
 })
@@ -240,8 +222,6 @@ export interface ParserState {
   readonly reasoningDetails: Array<unknown>
   readonly reasoningDetailsObserved: boolean
   readonly reasoningEmitted: boolean
-  readonly latestToolIndex?: number
-  readonly nextToolIndex: number
 }
 
 // =============================================================================
@@ -371,13 +351,12 @@ const lowerAssistantMessage = Effect.fn("OpenAIChat.lowerAssistantMessage")(func
     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 } : {}),
+    content: content.length === 0 ? null : ProviderShared.joinText(content),
+    tool_calls: toolCalls.length === 0 ? undefined : toolCalls,
+    reasoning_details: details,
+    cache_control: options.cacheControl?.(cached && "cache" in cached ? cached.cache : undefined),
   }
   if (field === undefined || reasoningText === undefined) return result
   return { ...result, [field]: reasoningText }
@@ -511,7 +490,6 @@ 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 } : {}),
   }
 }
@@ -581,34 +559,22 @@ const mapFinishReason = (reason: string | null | undefined): FinishReason => {
 // 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))
+  const cached = usage.prompt_tokens_details?.cached_tokens
+  const cacheWrite = usage.prompt_tokens_details?.cache_write_tokens
+  const reasoning = usage.completion_tokens_details?.reasoning_tokens
+  const nonCached = ProviderShared.subtractTokens(usage.prompt_tokens, ProviderShared.sumTokens(cached, cacheWrite))
   return new Usage({
-    inputTokens: input,
-    outputTokens: output,
+    inputTokens: usage.prompt_tokens,
+    outputTokens: usage.completion_tokens,
     nonCachedInputTokens: nonCached,
     cacheReadInputTokens: cached,
     cacheWriteInputTokens: cacheWrite,
     reasoningTokens: reasoning,
-    totalTokens: ProviderShared.totalTokens(input, output, usage.total_tokens ?? undefined),
+    totalTokens: ProviderShared.totalTokens(usage.prompt_tokens, usage.completion_tokens, usage.total_tokens),
     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,
@@ -691,39 +657,25 @@ const step = (state: ParserState, event: OpenAIChatEvent) =>
     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 finishReason = choice?.finish_reason
+      ? { normalized: mapFinishReason(choice.finish_reason), raw: choice.native_finish_reason ?? choice.finish_reason }
+      : 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 reasoningField = state.reasoningField ?? (!state.lifecycle.text.has("text-0") ? reasoning?.field : undefined)
     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)
+    if (!state.lifecycle.text.has("text-0") && text !== undefined)
+      lifecycle = Lifecycle.reasoningDelta(lifecycle, events, "reasoning-0", text, deltaMetadata)
     else if (
       reasoningDetailsObserved &&
       !lifecycle.reasoning.has("reasoning-0") &&
@@ -742,36 +694,24 @@ const step = (state: ParserState, event: OpenAIChatEvent) =>
       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]
+    for (const tool of toolDeltas) {
+      const current = tools[tool.index]
+      const pending = pendingTools[tool.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 },
-        }
+        pendingTools = { ...pendingTools, [tool.index]: { id: id || undefined, name: name || undefined, input: text } }
         continue
       }
       if (pending) {
         pendingTools = { ...pendingTools }
-        delete pendingTools[index]
+        delete pendingTools[tool.index]
       }
       const result = ToolStream.appendOrStart(
         ADAPTER,
         tools,
-        index,
+        tool.index,
         { id: id || undefined, name: name || undefined, text },
         "OpenAI Chat tool call delta is missing id or name",
       )
@@ -803,8 +743,6 @@ const step = (state: ParserState, event: OpenAIChatEvent) =>
         reasoningDetails: state.reasoningDetails,
         reasoningDetailsObserved,
         reasoningEmitted,
-        latestToolIndex,
-        nextToolIndex,
       },
       events,
     ] as const
@@ -812,18 +750,14 @@ const step = (state: ParserState, event: OpenAIChatEvent) =>
 
 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 hasToolCalls = state.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) }
+    : undefined
   const metadata = reasoningMetadata(
     state.reasoningField,
     state.reasoningDetailsObserved ? state.reasoningDetails : undefined,
@@ -833,9 +767,9 @@ const finishEvents = (state: ParserState): ReadonlyArray<LLMEvent> => {
       ? 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 })
+  const lifecycle = state.toolCallEvents.length ? Lifecycle.stepStart(ended, events) : ended
+  events.push(...state.toolCallEvents)
+  if (reason) Lifecycle.finish(lifecycle, events, { reason, usage: state.usage })
   return events
 }
 
@@ -865,7 +799,6 @@ export const protocol = Protocol.make({
       reasoningDetails: [],
       reasoningDetailsObserved: false,
       reasoningEmitted: false,
-      nextToolIndex: 0,
     }),
     step,
     onHalt: finishEvents,
@@ -884,4 +817,4 @@ export const route = Route.make({
   transport: httpTransport,
 })
 
-export * as OpenAIChat from "./openai-chat.js"
+export * as OpenAIChat from "./openai-chat"

packages/ai/src/protocols/openai-compatible-chat.ts

@@ -1,7 +1,7 @@
-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"
+import { Route, type RouteRoutedLanguageModelInput } from "../route/client"
+import { Endpoint } from "../route/endpoint"
+import { Framing } from "../route/framing"
+import * as OpenAIChat from "./openai-chat"
 
 const ADAPTER = "openai-compatible-chat"
 
@@ -22,4 +22,4 @@ export const route = Route.make({
   framing: Framing.sse,
 })
 
-export * as OpenAICompatibleChat from "./openai-compatible-chat.js"
+export * as OpenAICompatibleChat from "./openai-compatible-chat"

packages/ai/src/protocols/openai-compatible-responses.ts

@@ -1,6 +1,6 @@
-import { Route, type RouteRoutedLanguageModelInput } from "../route/client.js"
-import { Endpoint } from "../route/endpoint.js"
-import { OpenResponses } from "./open-responses.js"
+import { Route, type RouteRoutedLanguageModelInput } from "../route/client"
+import { Endpoint } from "../route/endpoint"
+import { OpenResponses } from "./open-responses"
 
 const ADAPTER = "openai-compatible-responses"
 
@@ -19,4 +19,4 @@ export const route = Route.make({
   transport: OpenResponses.httpTransport,
 })
 
-export * as OpenAICompatibleResponses from "./openai-compatible-responses.js"
+export * as OpenAICompatibleResponses from "./openai-compatible-responses"

packages/ai/src/protocols/openai-images.ts

@@ -7,8 +7,8 @@ import {
   type ImageInput,
   type ImageRequestFor,
   type ImageRoute,
-} from "../image.js"
-import { Auth, type Definition as AuthDefinition } from "../route/auth.js"
+} from "../image"
+import { Auth, type Definition as AuthDefinition } from "../route/auth"
 import {
   InvalidProviderOutputReason,
   AIError,
@@ -16,10 +16,10 @@ import {
   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"
+} from "../schema"
+import { ProviderShared } from "./shared"
+import { ImageInputs } from "./utils/image-input"
+import { OpenAIImage } from "./utils/openai-image"
 
 const ADAPTER = "openai-images"
 export const DEFAULT_BASE_URL = "https://api.openai.com/v1"

packages/ai/src/protocols/openai-responses.ts

@@ -1,18 +1,22 @@
 import { Effect, Encoding, Schema } from "effect"
-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, WebSocketTransport } 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 { Headers } from "effect/unstable/http"
+import { Route } from "../route/client"
+import { Auth } from "../route/auth"
+import { Endpoint } from "../route/endpoint"
+import { Protocol } from "../route/protocol"
+import { HttpTransport } from "../route/transport"
+import { LLMEvent, LLMRequest, type JsonSchema, type ToolDefinition } from "../schema"
+import { OpenResponses } from "./open-responses"
+import { optionalArray, ProviderShared } from "./shared"
+import { Lifecycle } from "./utils/lifecycle"
+import { OpenAIImage } from "./utils/openai-image"
+import { ToolSchemaProjection } from "./utils/tool-schema"
+import { OpenResponsesChannel } from "./open-responses-channel"
 
 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
 
@@ -57,16 +61,6 @@ const OpenAIResponsesBody = Schema.Struct({
 })
 export type OpenAIResponsesBody = Schema.Schema.Type<typeof OpenAIResponsesBody>
 
-const OpenAIResponsesWebSocketMessage = Schema.StructWithRest(
-  Schema.Struct({
-    type: Schema.tag("response.create"),
-    ...OpenAIResponsesCoreFields,
-  }),
-  [Schema.Record(Schema.String, Schema.Unknown)],
-)
-type OpenAIResponsesWebSocketMessage = Schema.Schema.Type<typeof OpenAIResponsesWebSocketMessage>
-const encodeWebSocketMessage = Schema.encodeSync(Schema.fromJsonString(OpenAIResponsesWebSocketMessage))
-
 const extension = {
   id: ADAPTER,
   name: NAME,
@@ -249,6 +243,12 @@ const endpoint = Endpoint.path<OpenAIResponsesBody>(PATH, { baseURL: DEFAULT_BAS
 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),
+})
 
 export const route = Route.make({
   id: ADAPTER,
@@ -257,37 +257,8 @@ export const route = Route.make({
   protocol,
   endpoint,
   auth,
-  transport: httpTransport,
+  transport,
   defaults: { providerOptions: { openai: { store: false } } },
 })
 
-const decodeWebSocketMessage = ProviderShared.validateWith(Schema.decodeUnknownEffect(OpenAIResponsesWebSocketMessage))
-
-const webSocketMessage = (body: OpenAIResponsesBody | Record<string, unknown>) =>
-  Effect.gen(function* () {
-    if (!ProviderShared.isRecord(body))
-      return yield* ProviderShared.invalidRequest("OpenAI Responses WebSocket body must be a JSON object")
-    const { stream: _stream, ...message } = body
-    return yield* decodeWebSocketMessage({ ...message, type: "response.create" })
-  })
-
-export const webSocketTransport = WebSocketTransport.jsonTransport.with<
-  OpenAIResponsesBody,
-  OpenAIResponsesWebSocketMessage
->({
-  toMessage: webSocketMessage,
-  encodeMessage: encodeWebSocketMessage,
-})
-
-export const webSocketRoute = Route.make({
-  id: `${ADAPTER}-websocket`,
-  provider: "openai",
-  providerMetadataKey: "openai",
-  protocol,
-  endpoint,
-  auth,
-  transport: webSocketTransport,
-  defaults: { providerOptions: { openai: { store: false } } },
-})
-
-export * as OpenAIResponses from "./openai-responses.js"
+export * as OpenAIResponses from "./openai-responses"

packages/ai/src/protocols/shared.ts

@@ -12,8 +12,8 @@ import {
   type MediaPart,
   type TextPart,
   type ToolResultPart,
-} from "../schema/index.js"
-import { isRecord } from "../utils/record.js"
+} from "../schema"
+import { isRecord } from "../utils/record"
 export { isRecord }
 
 export const Json = Schema.fromJsonString(Schema.Unknown)
@@ -324,4 +324,4 @@ export const jsonPost = (input: { readonly url: string; readonly body: string; r
     HttpClientRequest.bodyText(input.body, "application/json"),
   )
 
-export * as ProviderShared from "./shared.js"
+export * as ProviderShared from "./shared"

packages/ai/src/protocols/utils/bedrock-auth.ts

@@ -1,8 +1,8 @@
 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"
+import { Auth, type AuthInput } from "../../route/auth"
+import { ProviderShared } from "../shared"
 
 /**
  * AWS credentials for SigV4 signing. Bedrock also supports Bearer API key auth,
@@ -74,4 +74,4 @@ export const sigV4 = (
 /** Bedrock route auth defaults to SigV4 and expects credentials from route configuration. */
 export const auth = sigV4(undefined)
 
-export * as BedrockAuth from "./bedrock-auth.js"
+export * as BedrockAuth from "./bedrock-auth"

packages/ai/src/protocols/utils/bedrock-cache.ts

@@ -1,6 +1,6 @@
 import { Schema } from "effect"
-import type { CacheHint } from "../../schema/index.js"
-import { newBreakpoints, ttlBucket, type Breakpoints } from "./cache.js"
+import type { CacheHint } from "../../schema"
+import { newBreakpoints, ttlBucket, type Breakpoints } from "./cache"
 
 // Bedrock cache markers are positional: emit a `cachePoint` block immediately
 // after the content the caller wants treated as a cacheable prefix. Bedrock
@@ -18,7 +18,7 @@ export type CachePointBlock = Schema.Schema.Type<typeof CachePointBlock>
 // budget is respected across `system`, `messages`, and `tools`.
 export const BEDROCK_BREAKPOINT_CAP = 4
 
-export type { Breakpoints } from "./cache.js"
+export type { Breakpoints } from "./cache"
 export const breakpoints = () => newBreakpoints(BEDROCK_BREAKPOINT_CAP)
 
 const DEFAULT_5M: CachePointBlock = { cachePoint: { type: "default" } }
@@ -34,4 +34,4 @@ export const block = (breakpoints: Breakpoints, cache: CacheHint | undefined): C
   return ttlBucket(cache.ttlSeconds) === "1h" ? DEFAULT_1H : DEFAULT_5M
 }
 
-export * as BedrockCache from "./bedrock-cache.js"
+export * as BedrockCache from "./bedrock-cache"

packages/ai/src/protocols/utils/bedrock-media.ts

@@ -1,6 +1,6 @@
 import { Effect, Schema } from "effect"
-import type { MediaPart } from "../../schema/index.js"
-import { ProviderShared } from "../shared.js"
+import type { MediaPart } from "../../schema"
+import { ProviderShared } from "../shared"
 
 // Bedrock Converse accepts image `format` as the file extension and
 // `source.bytes` as base64 in the JSON wire format.
@@ -89,4 +89,4 @@ export const lower = Effect.fn("BedrockMedia.lower")(function* (part: MediaPart)
   return yield* ProviderShared.invalidRequest(`Bedrock Converse does not support media type ${part.mediaType}`)
 })
 
-export * as BedrockMedia from "./bedrock-media.js"
+export * as BedrockMedia from "./bedrock-media"

packages/ai/src/protocols/utils/gemini-tool-schema.ts

@@ -1,4 +1,4 @@
-import { isRecord } from "../../utils/record.js"
+import { isRecord } from "../../utils/record"
 
 // 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
@@ -61,59 +61,39 @@ const emptyObjectSchema = (schema: Record<string, unknown>) =>
   (!isRecord(schema.properties) || Object.keys(schema.properties).length === 0) &&
   !schema.additionalProperties
 
-const projectNode = (schema: unknown, nested = false): Record<string, unknown> | undefined => {
+const projectNode = (schema: unknown): 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(
+  if (emptyObjectSchema(schema)) return undefined
+  return 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,
-      ],
+      ["type", Array.isArray(schema.type) ? schema.type.filter((type) => type !== "null")[0] : schema.type],
+      ["nullable", Array.isArray(schema.type) && schema.type.includes("null") ? 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)]))
+          ? Object.fromEntries(Object.entries(schema.properties).map(([key, value]) => [key, projectNode(value)]))
           : undefined,
       ],
       [
         "items",
         Array.isArray(schema.items)
-          ? schema.items.map((item) => projectNode(item, true))
+          ? schema.items.map(projectNode)
           : schema.items === undefined
             ? undefined
-            : projectNode(schema.items, true),
+            : projectNode(schema.items),
       ],
-      ["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],
+      ["allOf", Array.isArray(schema.allOf) ? schema.allOf.map(projectNode) : undefined],
+      ["anyOf", Array.isArray(schema.anyOf) ? schema.anyOf.map(projectNode) : undefined],
+      ["oneOf", Array.isArray(schema.oneOf) ? schema.oneOf.map(projectNode) : 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"
+export * as GeminiToolSchema from "./gemini-tool-schema"

packages/ai/src/protocols/utils/image-input.ts

@@ -1,6 +1,6 @@
 import { Effect, Encoding } from "effect"
-import type { ImageInput } from "../../image.js"
-import { InvalidRequestReason, AIError } from "../../schema/index.js"
+import type { ImageInput } from "../../image"
+import { InvalidRequestReason, AIError } from "../../schema"
 
 const invalid = (module: string, message: string) =>
   new AIError({

packages/ai/src/protocols/utils/lifecycle.ts

@@ -1,4 +1,4 @@
-import { LLMEvent, type FinishReasonDetails, type ProviderMetadata, type Usage } from "../../schema/index.js"
+import { LLMEvent, type FinishReasonDetails, type ProviderMetadata, type Usage } from "../../schema"
 
 export interface State {
   readonly stepStarted: boolean
@@ -102,4 +102,4 @@ export const finish = (
   return { ...stepped, stepStarted: false }
 }
 
-export * as Lifecycle from "./lifecycle.js"
+export * as Lifecycle from "./lifecycle"

packages/ai/src/protocols/utils/open-responses-options.ts

@@ -1,5 +1,5 @@
 import { Schema } from "effect"
-import { TextVerbosity, type LLMRequest } from "../../schema/index.js"
+import { TextVerbosity, type LLMRequest } from "../../schema"
 
 export const ResponseIncludables = [
   "file_search_call.results",
@@ -33,6 +33,7 @@ export const ServiceTierSchema = Schema.Literals(ServiceTiers)
 export interface Resolved {
   readonly instructions?: string
   readonly store?: boolean
+  readonly promptCacheKey?: string
   readonly reasoningEffort?: string
   readonly reasoningSummary?: "auto" | "concise" | "detailed"
   readonly include?: ReadonlyArray<ResponseIncludable>
@@ -49,6 +50,7 @@ export const resolve = (request: LLMRequest): Resolved => {
   return {
     instructions: typeof input?.instructions === "string" ? input.instructions : undefined,
     store: typeof input?.store === "boolean" ? input.store : undefined,
+    promptCacheKey: typeof input?.promptCacheKey === "string" ? input.promptCacheKey : undefined,
     reasoningEffort: typeof input?.reasoningEffort === "string" ? input.reasoningEffort : undefined,
     reasoningSummary:
       reasoningSummary === "auto" || reasoningSummary === "concise" || reasoningSummary === "detailed"
@@ -60,4 +62,4 @@ export const resolve = (request: LLMRequest): Resolved => {
   }
 }
 
-export * as OpenResponsesOptions from "./open-responses-options.js"
+export * as OpenResponsesOptions from "./open-responses-options"

packages/ai/src/protocols/utils/openai-options.ts

@@ -1,5 +1,5 @@
-import { ReasoningEfforts } from "../../schema/index.js"
-import { OpenResponsesOptions } from "./open-responses-options.js"
+import { ReasoningEfforts } from "../../schema"
+import { OpenResponsesOptions } from "./open-responses-options"
 
 export const OpenAIReasoningEfforts = ReasoningEfforts
 export type OpenAIReasoningEffort = string
@@ -20,4 +20,4 @@ export const isReasoningEffort = (effort: unknown): effort is OpenAIReasoningEff
 
 export const resolve = OpenResponsesOptions.resolve
 
-export * as OpenAIOptions from "./openai-options.js"
+export * as OpenAIOptions from "./openai-options"

packages/ai/src/protocols/utils/tool-schema.ts

@@ -1,6 +1,6 @@
-import type { JsonSchema, LanguageModelToolSchemaCompatibility } from "../../schema/index.js"
-import { isRecord } from "../../utils/record.js"
-import { GeminiToolSchema } from "./gemini-tool-schema.js"
+import type { JsonSchema, LanguageModelToolSchemaCompatibility } from "../../schema"
+import { isRecord } from "../../utils/record"
+import { GeminiToolSchema } from "./gemini-tool-schema"
 
 const removeNullSchemas = (value: unknown): unknown => {
   if (Array.isArray(value)) return value.map(removeNullSchemas)

packages/ai/src/protocols/utils/tool-stream.ts

@@ -1,6 +1,6 @@
 import { Effect } from "effect"
-import { AIError, LLMEvent, type ProviderMetadata, type ToolCall, type ToolInputError } from "../../schema/index.js"
-import { eventError, parseToolInput, type ToolAccumulator } from "../shared.js"
+import { AIError, LLMEvent, type ProviderMetadata, type ToolCall, type ToolInputError } from "../../schema"
+import { eventError, parseToolInput, type ToolAccumulator } from "../shared"
 
 type StreamKey = string | number
 
@@ -223,4 +223,4 @@ export const finishAll = <K extends StreamKey>(route: string, tools: State<K>) =
     }
   })
 
-export * as ToolStream from "./tool-stream.js"
+export * as ToolStream from "./tool-stream"

packages/ai/src/protocols/xai-images.ts

@@ -1,7 +1,7 @@
 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 { GeneratedImage, ImageModel, ImageResponse, type ImageRequestFor, type ImageRoute } from "../image"
+import { Auth, type Definition as AuthDefinition } from "../route/auth"
 import {
   InvalidProviderOutputReason,
   AIError,
@@ -9,9 +9,9 @@ import {
   mergeHttpOptions,
   mergeJsonRecords,
   type HttpOptions,
-} from "../schema/index.js"
-import { ProviderShared, optionalNull } from "./shared.js"
-import { ImageInputs } from "./utils/image-input.js"
+} from "../schema"
+import { ProviderShared, optionalNull } from "./shared"
+import { ImageInputs } from "./utils/image-input"
 
 const ADAPTER = "xai-images"
 export const DEFAULT_BASE_URL = "https://api.x.ai/v1"

packages/ai/src/protocols/zai-images.ts

@@ -1,16 +1,10 @@
 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"
+import { GeneratedImage, ImageModel, ImageResponse, type ImageRequestFor, type ImageRoute } from "../image"
+import { Auth, type Definition as AuthDefinition } from "../route/auth"
+import { InvalidProviderOutputReason, AIError, mergeHttpOptions, mergeJsonRecords, type HttpOptions } from "../schema"
+import { ProviderShared } from "./shared"
+import { ImageInputs } from "./utils/image-input"
 
 const ADAPTER = "zai-images"
 export const DEFAULT_BASE_URL = "https://api.z.ai/api/paas/v4"

packages/ai/src/provider-error.ts

@@ -12,7 +12,7 @@ import {
   type HttpContext,
   type HttpRateLimitDetails,
   type ProviderMetadata,
-} from "./schema/index.js"
+} from "./schema"
 
 const patterns = [
   /prompt is too long/i,
@@ -67,6 +67,7 @@ const SERVER_CODES = new Set([
   "overloaded_error",
   "server_error",
   "server_is_overloaded",
+  "slow_down",
   "serviceunavailableexception",
 ])
 const INVALID_REQUEST_CODES = new Set(["invalid_prompt", "invalid_request_error", "validationexception"])

packages/ai/src/provider-package.ts

@@ -1,4 +1,4 @@
-import type { LanguageModel, ProviderOptions } from "./schema/index.js"
+import type { LanguageModel, ProviderOptions } from "./schema"
 
 export interface Settings extends Readonly<Record<string, unknown>> {
   readonly baseURL?: string
@@ -18,4 +18,4 @@ export interface Definition<
   readonly model: (modelID: string, settings: ProviderSettings) => LanguageModel<Options>
 }
 
-export * as ProviderPackage from "./provider-package.js"
+export * as ProviderPackage from "./provider-package"

packages/ai/src/provider.ts

@@ -1,4 +1,4 @@
-import type { LanguageModel, ModelID, ProviderID } from "./schema/index.js"
+import type { LanguageModel, ModelID, ProviderID } from "./schema"
 
 export type LanguageModelOptions = Pick<LanguageModel.Input, "defaults" | "compatibility">
 
@@ -33,4 +33,4 @@ export const make = <DefinitionType extends DefinitionShape>(
   definition: NoExtraFields<DefinitionType, DefinitionShape>,
 ) => definition
 
-export * as Provider from "./provider.js"
+export * as Provider from "./provider"

packages/ai/src/providers.ts

@@ -1 +1 @@
-export * from "./providers/index.js"
+export * from "./providers/index"

packages/ai/src/providers/amazon-bedrock-mantle.ts

@@ -1,11 +1,11 @@
-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"
+import { Auth } from "../route/auth"
+import type { Route as RouteDef, RouteDefaultsInput } from "../route/client"
+import type { ProviderPackage } from "../provider-package"
+import { OpenAIChat } from "../protocols/openai-chat"
+import { OpenAIResponses } from "../protocols/openai-responses"
+import { BedrockAuth, type Credentials } from "../protocols/utils/bedrock-auth"
+import { ProviderID, type ModelID } from "../schema"
+import { withOpenAIOptions, type OpenAIProviderOptionsInput } from "./openai-options"
 
 export const id = ProviderID.make("amazon-bedrock")
 

packages/ai/src/providers/amazon-bedrock.ts

@@ -1,9 +1,9 @@
-import type { RouteDefaultsInput } from "../route/client.js"
-import { Auth } from "../route/auth.js"
-import type { ProviderPackage } from "../provider-package.js"
-import { ProviderID, type ModelID } from "../schema/index.js"
-import * as BedrockConverse from "../protocols/bedrock-converse.js"
-import type { BedrockCredentials } from "../protocols/bedrock-converse.js"
+import type { RouteDefaultsInput } from "../route/client"
+import { Auth } from "../route/auth"
+import type { ProviderPackage } from "../provider-package"
+import { ProviderID, type ModelID } from "../schema"
+import * as BedrockConverse from "../protocols/bedrock-converse"
+import type { BedrockCredentials } from "../protocols/bedrock-converse"
 
 export const id = ProviderID.make("amazon-bedrock")
 

packages/ai/src/providers/amazon-bedrock/mantle.ts

@@ -1,2 +1,2 @@
-export { chatModel as model } from "../amazon-bedrock-mantle.js"
-export type { Settings } from "../amazon-bedrock-mantle.js"
+export { chatModel as model } from "../amazon-bedrock-mantle"
+export type { Settings } from "../amazon-bedrock-mantle"

packages/ai/src/providers/amazon-bedrock/mantle/chat.ts

@@ -1,2 +1,2 @@
-export { chatModel as model } from "../../amazon-bedrock-mantle.js"
-export type { Settings } from "../../amazon-bedrock-mantle.js"
+export { chatModel as model } from "../../amazon-bedrock-mantle"
+export type { Settings } from "../../amazon-bedrock-mantle"

packages/ai/src/providers/amazon-bedrock/mantle/responses.ts

@@ -1,2 +1,2 @@
-export { responsesModel as model } from "../../amazon-bedrock-mantle.js"
-export type { Settings } from "../../amazon-bedrock-mantle.js"
+export { responsesModel as model } from "../../amazon-bedrock-mantle"
+export type { Settings } from "../../amazon-bedrock-mantle"

packages/ai/src/providers/anthropic-compatible.ts

@@ -1,9 +1,9 @@
-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"
+import type { ProviderPackage } from "../provider-package"
+import { AnthropicMessages } from "../protocols/anthropic-messages"
+import { Auth } from "../route/auth"
+import type { ProviderAuthOption } from "../route/auth-options"
+import type { RouteDefaultsInput } from "../route/client"
+import { ProviderID, type ModelID } from "../schema"
 
 export type AnthropicOptionsInput = AnthropicMessages.OptionsInput
 export type AnthropicProviderOptionsInput = AnthropicMessages.ProviderOptionsInput
@@ -74,4 +74,4 @@ export const model: ProviderPackage.Definition<Settings, AnthropicMessages.Provi
   }).model(modelID)
 }
 
-export * as AnthropicCompatible from "./anthropic-compatible.js"
+export * as AnthropicCompatible from "./anthropic-compatible"

packages/ai/src/providers/anthropic.ts

@@ -1,10 +1,10 @@
-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"
+import type { RouteDefaultsInput } from "../route/client"
+import { Auth } from "../route/auth"
+import type { ProviderAuthOption } from "../route/auth-options"
+import type { ProviderPackage } from "../provider-package"
+import { ProviderID, type ModelID } from "../schema"
+import { AnthropicMessages } from "../protocols/anthropic-messages"
+import { AnthropicCompatible } from "./anthropic-compatible"
 
 export type AnthropicOptionsInput = AnthropicMessages.OptionsInput
 export type AnthropicProviderOptionsInput = AnthropicMessages.ProviderOptionsInput

packages/ai/src/providers/azure.ts

@@ -1,12 +1,11 @@
-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"
+import { Auth } from "../route/auth"
+import { type AtLeastOne, type ProviderAuthOption } from "../route/auth-options"
+import type { Route as RouteDef, RouteDefaultsInput } from "../route/client"
+import type { ProviderPackage } from "../provider-package"
+import { ProviderID, type ModelID } from "../schema"
+import * as OpenAIChat from "../protocols/openai-chat"
+import * as OpenAIResponses from "../protocols/openai-responses"
+import { withOpenAIOptions, type OpenAIProviderOptionsInput } from "./openai-options"
 
 export const id = ProviderID.make("azure")
 const routeAuth = Auth.remove("authorization")
@@ -20,7 +19,7 @@ export type LanguageModelOptions = AzureURL &
   ProviderAuthOption<"optional"> & {
     readonly apiVersion?: string
     readonly queryParams?: Record<string, string>
-    readonly useDeploymentBasedUrls?: boolean
+    readonly useCompletionUrls?: boolean
     readonly providerOptions?: OpenAIProviderOptionsInput
   }
 export type Config = LanguageModelOptions
@@ -30,22 +29,27 @@ export type Settings = ProviderPackage.Settings &
     readonly apiKey?: string
     readonly apiVersion?: string
     readonly queryParams?: Readonly<Record<string, string>>
-    readonly useDeploymentBasedUrls?: boolean
     readonly providerOptions?: OpenAIProviderOptionsInput
   }
 
-const resourceBaseURL = (resourceName: string) => `https://${resourceName.trim()}.openai.azure.com/openai`
+const resourceBaseURL = (resourceName: string) => `https://${resourceName.trim()}.openai.azure.com/openai/v1`
 
 const responsesRoute = OpenAIResponses.route.with({
   id: "azure-openai-responses",
   provider: id,
   auth: routeAuth,
+  endpoint: {
+    query: { "api-version": "v1" },
+  },
 })
 
 const chatRoute = OpenAIChat.route.with({
   id: "azure-openai-chat",
   provider: id,
   auth: routeAuth,
+  endpoint: {
+    query: { "api-version": "v1" },
+  },
 })
 
 export const routes = [responsesRoute, chatRoute]
@@ -55,7 +59,7 @@ const defaults = (input: Config) => {
     apiKey: _,
     apiVersion: _apiVersion,
     resourceName: _resourceName,
-    useDeploymentBasedUrls: _useDeploymentBasedUrls,
+    useCompletionUrls: _useCompletionUrls,
     baseURL: _baseURL,
     queryParams: _queryParams,
     ...rest
@@ -76,39 +80,37 @@ const auth = (input: Config) => {
   )
 }
 
-const configuredRoute = <Body, Prepared>(route: RouteDef<Body, Prepared>, input: Config, modelID: string | ModelID) =>
+const configuredRoute = <Body, Prepared>(route: RouteDef<Body, Prepared>, input: Config) =>
   route.with({
     auth: auth(input),
-    endpoint: endpoint(input, modelID),
+    endpoint: {
+      // AtLeastOne guarantees at least one is set; baseURL wins if both are.
+      baseURL: input.baseURL ?? resourceBaseURL(input.resourceName!),
+      query: {
+        ...(input.apiVersion ? { "api-version": input.apiVersion } : {}),
+        ...input.queryParams,
+      },
+    },
   })
 
-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 configuredResponsesRoute = configuredRoute(responsesRoute, input)
+  const configuredChatRoute = configuredRoute(chatRoute, input)
   const modelDefaults = defaults(input)
 
   const responses = (modelID: string | ModelID) =>
-    configuredRoute(responsesRoute, input, modelID)
+    configuredResponsesRoute
       .with(withOpenAIOptions(modelID, modelDefaults))
       .model<OpenAIProviderOptionsInput>({ id: modelID })
 
   const chat = (modelID: string | ModelID) =>
-    configuredRoute(chatRoute, input, modelID)
+    configuredChatRoute
       .with(withOpenAIOptions(modelID, modelDefaults))
       .model<OpenAIProviderOptionsInput>({ id: modelID })
 
   return {
     id,
-    model: responses,
+    model: (modelID: string | ModelID) => (input.useCompletionUrls === true ? chat(modelID) : responses(modelID)),
     responses,
     chat,
     configure,
@@ -129,7 +131,6 @@ const config = (settings: Settings): Config => {
     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 }

packages/ai/src/providers/azure/chat.ts

@@ -1,2 +1,2 @@
-export { chatModel as model } from "../azure.js"
-export type { Settings } from "../azure.js"
+export { chatModel as model } from "../azure"
+export type { Settings } from "../azure"

packages/ai/src/providers/azure/responses.ts

@@ -1,2 +1,2 @@
-export { responsesModel as model } from "../azure.js"
-export type { Settings } from "../azure.js"
+export { responsesModel as model } from "../azure"
+export type { Settings } from "../azure"

packages/ai/src/providers/cloudflare.ts

@@ -1,10 +1,10 @@
 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"
+import * as OpenAICompatibleChat from "../protocols/openai-compatible-chat"
+import { Auth } from "../route/auth"
+import { AuthOptions, type AtLeastOne, type ProviderAuthOption } from "../route/auth-options"
+import type { RouteDefaultsInput } from "../route/client"
+import { ProviderID, type ModelID } from "../schema"
+import type { OpenAIProviderOptionsInput } from "./openai-options"
 
 export const aiGatewayID = ProviderID.make("cloudflare-ai-gateway")
 export const workersAIID = ProviderID.make("cloudflare-workers-ai")

packages/ai/src/providers/github-copilot.ts

@@ -0,0 +1,71 @@
+import { AuthOptions, type ProviderAuthOption } from "../route/auth-options"
+import type { RouteDefaultsInput } from "../route/client"
+import { ProviderID, type ModelID } from "../schema"
+import * as OpenAIChat from "../protocols/openai-chat"
+import * as OpenAIResponses from "../protocols/openai-responses"
+import { withOpenAIOptions, type OpenAIProviderOptionsInput } from "./openai-options"
+
+export const id = ProviderID.make("github-copilot")
+
+// GitHub Copilot has no canonical public URL — callers (opencode, etc.) must
+// supply `baseURL` explicitly.
+export type LanguageModelOptions = Omit<RouteDefaultsInput, "providerOptions"> &
+  ProviderAuthOption<"optional"> & {
+    readonly baseURL: string
+    readonly endpoint?: "chat" | "responses"
+    readonly providerOptions?: OpenAIProviderOptionsInput
+  }
+
+export const shouldUseResponsesApi = (modelID: string | ModelID, endpoint?: LanguageModelOptions["endpoint"]) => {
+  if (endpoint) return endpoint === "responses"
+  const model = String(modelID)
+  const match = /^gpt-(\d+)/.exec(model)
+  if (!match) return false
+  return Number(match[1]) >= 5 && !model.startsWith("gpt-5-mini")
+}
+
+export const routes = [OpenAIResponses.route, OpenAIChat.route]
+
+const chatRoute = OpenAIChat.route.with({ provider: id })
+const responsesRoute = OpenAIResponses.route.with({ provider: id })
+
+const defaults = (options: LanguageModelOptions) => {
+  const { apiKey: _, auth: _auth, baseURL: _baseURL, endpoint: _endpoint, ...rest } = options
+  return rest
+}
+
+const configuredResponsesRoute = (options: LanguageModelOptions) =>
+  responsesRoute.with({
+    endpoint: { baseURL: options.baseURL },
+    auth: AuthOptions.bearer(options, []),
+  })
+
+const configuredChatRoute = (options: LanguageModelOptions) =>
+  chatRoute.with({
+    endpoint: { baseURL: options.baseURL },
+    auth: AuthOptions.bearer(options, []),
+  })
+
+export const configure = (options: LanguageModelOptions) => {
+  const responsesRoute = configuredResponsesRoute(options)
+  const chatRoute = configuredChatRoute(options)
+  const responses = (modelID: string | ModelID) =>
+    responsesRoute
+      .with(withOpenAIOptions(modelID, defaults(options)))
+      .model<OpenAIProviderOptionsInput>({ id: modelID })
+  const chat = (modelID: string | ModelID) =>
+    chatRoute.with(withOpenAIOptions(modelID, defaults(options))).model<OpenAIProviderOptionsInput>({ id: modelID })
+  return {
+    id,
+    model: (modelID: string | ModelID) =>
+      shouldUseResponsesApi(modelID, options.endpoint) ? responses(modelID) : chat(modelID),
+    responses,
+    chat,
+    configure,
+  }
+}
+
+export const provider = {
+  id,
+  configure,
+}

packages/ai/src/providers/google-vertex-chat.ts

@@ -1,9 +1,9 @@
-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"
+import type { ProviderPackage } from "../provider-package"
+import { OpenAICompatibleChat } from "../protocols/openai-compatible-chat"
+import type { RouteDefaultsInput } from "../route/client"
+import { ProviderID, type ModelID } from "../schema"
+import { GoogleVertexShared } from "./google-vertex-shared"
+import type { OpenAIProviderOptionsInput } from "./openai-options"
 
 export const id = ProviderID.make("google-vertex")
 

packages/ai/src/providers/google-vertex-messages.ts

@@ -1,13 +1,13 @@
 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"
+import type { ProviderPackage } from "../provider-package"
+import { AnthropicMessages } from "../protocols/anthropic-messages"
+import { Auth } from "../route/auth"
+import { Route, type RouteDefaultsInput } from "../route/client"
+import { Endpoint } from "../route/endpoint"
+import { Framing } from "../route/framing"
+import { Protocol } from "../route/protocol"
+import { ProviderID, type ModelID } from "../schema"
+import { GoogleVertexShared } from "./google-vertex-shared"
 
 export type AnthropicOptionsInput = AnthropicMessages.OptionsInput
 export type AnthropicProviderOptionsInput = AnthropicMessages.ProviderOptionsInput
@@ -15,7 +15,8 @@ export type AnthropicThinkingInput = AnthropicMessages.ThinkingInput
 
 const VERSION = "vertex-2023-10-16" as const
 
-export const id = ProviderID.make("google-vertex")
+// models.dev uses this provider id even though the API contract is Anthropic Messages.
+export const id = ProviderID.make("google-vertex-anthropic")
 
 export type Config = RouteDefaultsInput &
   GoogleVertexShared.OAuthOptions & {

packages/ai/src/providers/google-vertex-responses.ts

@@ -1,9 +1,9 @@
-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"
+import type { ProviderPackage } from "../provider-package"
+import { OpenAICompatibleResponses } from "../protocols/openai-compatible-responses"
+import type { RouteDefaultsInput } from "../route/client"
+import { ProviderID, type ModelID } from "../schema"
+import { GoogleVertexShared } from "./google-vertex-shared"
+import type { OpenResponsesProviderOptionsInput } from "./open-responses-options"
 
 export const id = ProviderID.make("google-vertex")
 

packages/ai/src/providers/google-vertex-shared.ts

@@ -1,6 +1,6 @@
 import type { AnyAuthClient } from "google-auth-library"
 import { Effect, Redacted } from "effect"
-import { Auth, MissingCredentialError } from "../route/auth.js"
+import { Auth, MissingCredentialError } from "../route/auth"
 
 const SCOPE = "https://www.googleapis.com/auth/cloud-platform"
 
@@ -74,4 +74,4 @@ export const oauth = (input: OAuthOptions, project?: string) => {
   return adc(project)
 }
 
-export * as GoogleVertexShared from "./google-vertex-shared.js"
+export * as GoogleVertexShared from "./google-vertex-shared"

packages/ai/src/providers/google-vertex.ts

@@ -1,11 +1,11 @@
-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"
+import type { ProviderPackage } from "../provider-package"
+import { Gemini } from "../protocols/gemini"
+import { Auth } from "../route/auth"
+import { Route, type RouteDefaultsInput } from "../route/client"
+import { Endpoint } from "../route/endpoint"
+import { Framing } from "../route/framing"
+import { ProviderID, type ModelID } from "../schema"
+import { GoogleVertexShared } from "./google-vertex-shared"
 
 export type GeminiOptionsInput = Gemini.OptionsInput
 export type GeminiProviderOptionsInput = Gemini.ProviderOptionsInput

packages/ai/src/providers/google-vertex/chat.ts

@@ -1,2 +1,2 @@
-export { model } from "../google-vertex-chat.js"
-export type { Settings } from "../google-vertex-chat.js"
+export { model } from "../google-vertex-chat"
+export type { Settings } from "../google-vertex-chat"

packages/ai/src/providers/google-vertex/gemini.ts

@@ -1,2 +1,2 @@
-export { model } from "../google-vertex.js"
-export type { Settings } from "../google-vertex.js"
+export { model } from "../google-vertex"
+export type { Settings } from "../google-vertex"

packages/ai/src/providers/google-vertex/messages.ts

@@ -1,2 +1,2 @@
-export { model } from "../google-vertex-messages.js"
-export type { Settings } from "../google-vertex-messages.js"
+export { model } from "../google-vertex-messages"
+export type { Settings } from "../google-vertex-messages"

packages/ai/src/providers/google-vertex/responses.ts

@@ -1,2 +1,2 @@
-export { model } from "../google-vertex-responses.js"
-export type { Settings } from "../google-vertex-responses.js"
+export { model } from "../google-vertex-responses"
+export type { Settings } from "../google-vertex-responses"

packages/ai/src/providers/google.ts

@@ -1,12 +1,12 @@
-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"
+import type { RouteDefaultsInput } from "../route/client"
+import { Auth } from "../route/auth"
+import type { ProviderAuthOption } from "../route/auth-options"
+import type { ProviderPackage } from "../provider-package"
+import { HttpOptions, ProviderID, mergeHttpOptions, type ModelID } from "../schema"
+import { Gemini } from "../protocols/gemini"
+import { GoogleImages } from "../protocols/google-images"
 
-export type { GoogleImageOptions } from "../protocols/google-images.js"
+export type { GoogleImageOptions } from "../protocols/google-images"
 export type GeminiOptionsInput = Gemini.OptionsInput
 export type GeminiProviderOptionsInput = Gemini.ProviderOptionsInput
 

packages/ai/src/providers/index.ts

@@ -1,18 +1,19 @@
-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"
+export * as Anthropic from "./anthropic"
+export * as AnthropicCompatible from "./anthropic-compatible"
+export * as AmazonBedrock from "./amazon-bedrock"
+export * as AmazonBedrockMantle from "./amazon-bedrock-mantle"
+export * as Azure from "./azure"
+export * as Cloudflare from "./cloudflare"
+export { CloudflareAIGateway, CloudflareWorkersAI } from "./cloudflare"
+export * as GitHubCopilot from "./github-copilot"
+export * as Google from "./google"
+export * as GoogleVertex from "./google-vertex"
+export * as GoogleVertexChat from "./google-vertex-chat"
+export * as GoogleVertexMessages from "./google-vertex-messages"
+export * as GoogleVertexResponses from "./google-vertex-responses"
+export * as OpenAI from "./openai"
+export * as OpenAICompatible from "./openai-compatible"
+export * as OpenAICompatibleResponses from "./openai-compatible-responses"
+export * as OpenRouter from "./openrouter"
+export * as XAI from "./xai"
+export * as ZAI from "./zai"

packages/ai/src/providers/open-responses-options.ts

@@ -1,10 +1,11 @@
-import type { ResponseIncludable, ServiceTier } from "../protocols/utils/open-responses-options.js"
-import type { ProviderOptions, ReasoningEffort, TextVerbosity } from "../schema/index.js"
+import type { ResponseIncludable, ServiceTier } from "../protocols/utils/open-responses-options"
+import type { ProviderOptions, ReasoningEffort, TextVerbosity } from "../schema"
 
 export interface OpenResponsesOptionsInput {
   readonly [key: string]: unknown
   readonly instructions?: string
   readonly store?: boolean
+  readonly promptCacheKey?: string
   readonly reasoningEffort?: ReasoningEffort
   readonly reasoningSummary?: "auto" | "concise" | "detailed"
   readonly include?: ReadonlyArray<ResponseIncludable>
@@ -16,4 +17,4 @@ export type OpenResponsesProviderOptionsInput = ProviderOptions & {
   readonly openresponses?: OpenResponsesOptionsInput
 }
 
-export * as OpenResponsesProviderOptions from "./open-responses-options.js"
+export * as OpenResponsesProviderOptions from "./open-responses-options"

packages/ai/src/providers/openai-compatible-responses.ts

@@ -1,11 +1,11 @@
-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"
+import type { ProviderPackage } from "../provider-package"
+import { OpenAICompatibleResponses } from "../protocols/openai-compatible-responses"
+import { AuthOptions, type ProviderAuthOption } from "../route/auth-options"
+import type { RouteDefaultsInput } from "../route/client"
+import { ProviderID, type ModelID } from "../schema"
+import type { OpenResponsesProviderOptionsInput } from "./open-responses-options"
 
-export type { OpenResponsesOptionsInput, OpenResponsesProviderOptionsInput } from "./open-responses-options.js"
+export type { OpenResponsesOptionsInput, OpenResponsesProviderOptionsInput } from "./open-responses-options"
 
 export const id = ProviderID.make("openai-compatible")
 

packages/ai/src/providers/openai-compatible.ts

@@ -1,10 +1,10 @@
-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"
+import { ProviderID, type ModelID } from "../schema"
+import * as OpenAICompatibleChat from "../protocols/openai-compatible-chat"
+import type { RouteDefaultsInput } from "../route/client"
+import { AuthOptions, type ProviderAuthOption } from "../route/auth-options"
+import type { ProviderPackage } from "../provider-package"
+import { profiles, type OpenAICompatibleProfile } from "./openai-compatible-profile"
+import type { OpenAIProviderOptionsInput } from "./openai-options"
 
 export const id = ProviderID.make("openai-compatible")
 

packages/ai/src/providers/openai-compatible/responses.ts

@@ -1 +1 @@
-export * from "../openai-compatible-responses.js"
+export * from "../openai-compatible-responses"

packages/ai/src/providers/openai-options.ts

@@ -1,8 +1,8 @@
-import type { ProviderOptions } from "../schema/index.js"
-import { mergeProviderOptions } from "../schema/index.js"
-import type { OpenResponsesOptionsInput } from "./open-responses-options.js"
+import type { ProviderOptions } from "../schema"
+import { mergeProviderOptions } from "../schema"
+import type { OpenResponsesOptionsInput } from "./open-responses-options"
 
-export type { OpenAIResponseIncludable, OpenAIServiceTier } from "../protocols/utils/openai-options.js"
+export type { OpenAIResponseIncludable, OpenAIServiceTier } from "../protocols/utils/openai-options"
 
 export type OpenAIOptionsInput = OpenResponsesOptionsInput
 
@@ -17,6 +17,7 @@ const openAIProviderOptions = (options: OpenAIOptionsInput | undefined): Provide
   const openai = Object.fromEntries(
     definedEntries({
       store: options?.store,
+      promptCacheKey: options?.promptCacheKey,
       reasoningEffort: options?.reasoningEffort,
       reasoningSummary: options?.reasoningSummary,
       include: options?.include,
@@ -67,4 +68,4 @@ export const withOpenAIOptions = <Options extends { readonly providerOptions?: O
   }
 }
 
-export * as OpenAIProviderOptions from "./openai-options.js"
+export * as OpenAIProviderOptions from "./openai-options"

packages/ai/src/providers/openai.ts

@@ -1,18 +1,18 @@
-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"
+import { AuthOptions, type ProviderAuthOption } from "../route/auth-options"
+import type { Route, RouteDefaultsInput } from "../route/client"
+import type { ProviderPackage } from "../provider-package"
+import { HttpOptions, ProviderID, ToolDefinition, mergeHttpOptions, type ModelID } from "../schema"
+import * as OpenAIChat from "../protocols/openai-chat"
+import * as OpenAIResponses from "../protocols/openai-responses"
+import { withOpenAIOptions, type OpenAIProviderOptionsInput } from "./openai-options"
+import { OpenAIImages, type OpenAIImageString } from "../protocols/openai-images"
 
-export type { OpenAIOptionsInput, OpenAIResponseIncludable } from "./openai-options.js"
-export type { OpenAIImageOptions } from "../protocols/openai-images.js"
+export type { OpenAIOptionsInput, OpenAIResponseIncludable } from "./openai-options"
+export type { OpenAIImageOptions } from "../protocols/openai-images"
 
 export const id = ProviderID.make("openai")
 
-export const routes = [OpenAIResponses.route, OpenAIResponses.webSocketRoute, OpenAIChat.route]
+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,
@@ -63,7 +63,6 @@ export interface Settings extends ProviderPackage.Settings {
   readonly organization?: string
   readonly project?: string
   readonly queryParams?: Readonly<Record<string, string>>
-  readonly transport?: "http" | "websocket"
   readonly providerOptions?: OpenAIProviderOptionsInput
 }
 
@@ -82,17 +81,12 @@ const configuredRoute = <Body, Prepared>(route: Route<Body, Prepared>, input: Co
 
 export const configure = (input: Config = {}) => {
   const responsesRoute = configuredRoute(OpenAIResponses.route, input)
-  const responsesWebSocketRoute = configuredRoute(OpenAIResponses.webSocketRoute, 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 responsesWebSocket = (id: string | ModelID) =>
-    responsesWebSocketRoute
-      .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) =>
@@ -111,7 +105,6 @@ export const configure = (input: Config = {}) => {
     id,
     model: responses,
     responses,
-    responsesWebSocket,
     chat,
     image,
     configure,
@@ -138,10 +131,7 @@ const config = (settings: Settings): Config => {
 }
 
 export const model: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (modelID, settings) => {
-  const configured = configure(config(settings))
-  if (settings.transport === undefined || settings.transport === "http") return configured.responses(modelID)
-  if (settings.transport === "websocket") return configured.responsesWebSocket(modelID)
-  throw new Error(`Unsupported OpenAI Responses transport: ${String(settings.transport)}`)
+  return configure(config(settings)).responses(modelID)
 }
 
 export const chatModel: ProviderPackage.Definition<Settings, OpenAIProviderOptionsInput>["model"] = (
@@ -149,6 +139,5 @@ export const chatModel: ProviderPackage.Definition<Settings, OpenAIProviderOptio
   settings,
 ) => configure(config(settings)).chat(modelID)
 export const responses = provider.responses
-export const responsesWebSocket = provider.responsesWebSocket
 export const chat = provider.chat
 export const image = provider.image

packages/ai/src/providers/openai/chat.ts

@@ -1,2 +1,2 @@
-export { chatModel as model } from "../openai.js"
-export type { Settings } from "../openai.js"
+export { chatModel as model } from "../openai"
+export type { Settings } from "../openai"

packages/ai/src/providers/openai/responses.ts

@@ -1,2 +1,2 @@
-export { model } from "../openai.js"
-export type { Settings } from "../openai.js"
+export { model } from "../openai"
+export type { Settings } from "../openai"

packages/ai/src/providers/openrouter.ts

@@ -1,15 +1,15 @@
 import { Effect, Schema } from "effect"
-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 { AuthOptions, type ProviderAuthOption } from "../route/auth-options.js"
-import { ProviderID, type CacheHint, type ModelID, type ProviderOptions } from "../schema/index.js"
-import type { ProviderPackage } from "../provider-package.js"
-import * as OpenAICompatibleProfiles from "./openai-compatible-profile.js"
-import * as OpenAIChat from "../protocols/openai-chat.js"
-import { newBreakpoints, ttlBucket } from "../protocols/utils/cache.js"
-import { isRecord } from "../protocols/shared.js"
+import { Route, type RouteDefaultsInput } from "../route/client"
+import { Endpoint } from "../route/endpoint"
+import { Framing } from "../route/framing"
+import { Protocol } from "../route/protocol"
+import { AuthOptions, type ProviderAuthOption } from "../route/auth-options"
+import { ProviderID, type CacheHint, type ModelID, type ProviderOptions } from "../schema"
+import type { ProviderPackage } from "../provider-package"
+import * as OpenAICompatibleProfiles from "./openai-compatible-profile"
+import * as OpenAIChat from "../protocols/openai-chat"
+import { newBreakpoints, ttlBucket } from "../protocols/utils/cache"
+import { isRecord } from "../protocols/shared"
 
 export const profile = OpenAICompatibleProfiles.profiles.openrouter
 export const id = ProviderID.make(profile.provider)
@@ -55,6 +55,7 @@ export interface OpenRouterOptions {
   readonly debug?: Readonly<{ echo_upstream_body?: boolean }>
   readonly models?: ReadonlyArray<string>
   readonly plugins?: ReadonlyArray<OpenRouterPlugin>
+  readonly promptCacheKey?: string
   readonly provider?: OpenRouterProviderRouting
   readonly reasoning?: Readonly<{
     enabled?: boolean
@@ -121,7 +122,6 @@ export const protocol = Protocol.make({
             ...body,
             messages,
             ...bodyOptions(request.providerOptions?.openrouter),
-            ...(request.promptCacheKey ? { prompt_cache_key: request.promptCacheKey } : {}),
           } as OpenRouterBody
         }),
       ),
@@ -161,6 +161,7 @@ const bodyOptions = (input: unknown) => {
     ...(isRecord(debug) ? { debug } : {}),
     ...(typeof user === "string" ? { user } : {}),
     ...(isRecord(reasoning) ? { reasoning } : {}),
+    ...(typeof promptCacheKey === "string" ? { prompt_cache_key: promptCacheKey } : {}),
   }
 }
 

packages/ai/src/providers/xai.ts

@@ -1,14 +1,14 @@
-import { AuthOptions, type ProviderAuthOption } from "../route/auth-options.js"
-import { Route, type RouteDefaultsInput } from "../route/client.js"
-import { Endpoint } from "../route/endpoint.js"
-import { HttpOptions, ProviderID, type ModelID, type ProviderOptions } from "../schema/index.js"
-import * as OpenAICompatibleProfiles from "./openai-compatible-profile.js"
-import * as OpenAICompatibleChat from "../protocols/openai-compatible-chat.js"
-import * as OpenAIChat from "../protocols/openai-chat.js"
-import * as OpenAIResponses from "../protocols/openai-responses.js"
-import { XAIImages } from "../protocols/xai-images.js"
-import type { OpenAIOptionsInput } from "./openai-options.js"
-import type { ProviderPackage } from "../provider-package.js"
+import { AuthOptions, type ProviderAuthOption } from "../route/auth-options"
+import { Route, type RouteDefaultsInput } from "../route/client"
+import { Endpoint } from "../route/endpoint"
+import { HttpOptions, ProviderID, type ModelID, type ProviderOptions } from "../schema"
+import * as OpenAICompatibleProfiles from "./openai-compatible-profile"
+import * as OpenAICompatibleChat from "../protocols/openai-compatible-chat"
+import * as OpenAIChat from "../protocols/openai-chat"
+import * as OpenAIResponses from "../protocols/openai-responses"
+import { XAIImages } from "../protocols/xai-images"
+import type { OpenAIOptionsInput } from "./openai-options"
+import type { ProviderPackage } from "../provider-package"
 
 export const id = ProviderID.make("xai")
 
@@ -28,7 +28,7 @@ export interface Settings extends ProviderPackage.Settings {
   readonly providerOptions?: XAIProviderOptionsInput
 }
 
-export type { XAIImageOptions } from "../protocols/xai-images.js"
+export type { XAIImageOptions } from "../protocols/xai-images"
 
 const responsesRoute = Route.make({
   id: "openai-responses",
@@ -47,8 +47,6 @@ const chatRoute = Route.make({
   protocol: OpenAIChat.protocol,
   endpoint: Endpoint.path("/chat/completions", { baseURL: OpenAICompatibleProfiles.profiles.xai.baseURL }),
   transport: OpenAICompatibleChat.route.transport,
-  headers: ({ request }): Record<string, string> =>
-    request.promptCacheKey ? { "x-grok-conv-id": request.promptCacheKey } : {},
 })
 
 export const routes = [responsesRoute, chatRoute]

packages/ai/src/providers/zai.ts

@@ -1,6 +1,6 @@
-import { ZAIImages } from "../protocols/zai-images.js"
-import { AuthOptions, type ProviderAuthOption } from "../route/auth-options.js"
-import { HttpOptions, ProviderID, type ModelID } from "../schema/index.js"
+import { ZAIImages } from "../protocols/zai-images"
+import { AuthOptions, type ProviderAuthOption } from "../route/auth-options"
+import { HttpOptions, ProviderID, type ModelID } from "../schema"
 
 export const id = ProviderID.make("zai")
 
@@ -10,7 +10,7 @@ export type Config = ProviderAuthOption<"optional"> & {
   readonly http?: HttpOptions.Input
 }
 
-export type { ZAIImageOptions } from "../protocols/zai-images.js"
+export type { ZAIImageOptions } from "../protocols/zai-images"
 
 const auth = (options: ProviderAuthOption<"optional">) => AuthOptions.bearer(options, "ZAI_API_KEY")
 

packages/ai/src/route.ts

@@ -1 +1 @@
-export * from "./route/index.js"
+export * from "./route/index"

packages/ai/src/route/auth-options.ts

@@ -1,5 +1,5 @@
 import type { Config, Redacted } from "effect"
-import { Auth } from "./auth.js"
+import { Auth } from "./auth"
 
 export type ApiKeyMode = "optional" | "required"
 
@@ -61,4 +61,4 @@ export const bearer = (
     .bearer()
 }
 
-export * as AuthOptions from "./auth-options.js"
+export * as AuthOptions from "./auth-options"

packages/ai/src/route/auth.ts

@@ -1,6 +1,6 @@
 import { Config, Effect, Redacted } from "effect"
 import { Headers } from "effect/unstable/http"
-import { AuthenticationReason, InvalidRequestReason, AIError, type HttpOptions } from "../schema/index.js"
+import { AuthenticationReason, InvalidRequestReason, AIError, type HttpOptions } from "../schema"
 
 export class MissingCredentialError extends Error {
   readonly _tag = "MissingCredentialError"
@@ -153,4 +153,4 @@ export const toEffect =
   (authInput: AuthInput): Effect.Effect<Headers.Headers, AIError> =>
     input.apply(authInput).pipe(Effect.mapError(toAIError))
 
-export * as Auth from "./auth.js"
+export * as Auth from "./auth"

packages/ai/src/route/client.ts

@@ -1,16 +1,14 @@
 import { Cause, Context, Effect, Layer, Schema, Stream } from "effect"
-import * as Option from "effect/Option"
-import { Auth } from "./auth.js"
-import { Endpoint, type EndpointPatch } from "./endpoint.js"
-import { RequestExecutor } from "./executor.js"
-import { Framing } from "./framing.js"
-import { HttpTransport } from "./transport/index.js"
-import type { HttpMiddleware, Transport, TransportRuntime } from "./transport/index.js"
-import { WebSocketExecutor } from "./transport/index.js"
-import type { Protocol } from "./protocol.js"
-import { applyCachePolicy } from "../cache-policy.js"
-import * as ProviderShared from "../protocols/shared.js"
-import type { ProtocolID, ProviderOptions } from "../schema/index.js"
+import { Auth } from "./auth"
+import { Endpoint, type EndpointPatch } from "./endpoint"
+import { RequestExecutor } from "./executor"
+import { Framing } from "./framing"
+import { HttpTransport } from "./transport"
+import type { HttpMiddleware, Transport, TransportRuntime, WebSocketChannelExecutor } from "./transport"
+import type { Protocol } from "./protocol"
+import { applyCachePolicy } from "../cache-policy"
+import * as ProviderShared from "../protocols/shared"
+import type { ProtocolID, ProviderOptions } from "../schema"
 import {
   AIError,
   GenerationOptions,
@@ -25,7 +23,7 @@ import {
   mergeGenerationOptions,
   mergeHttpOptions,
   mergeProviderOptions,
-} from "../schema/index.js"
+} from "../schema"
 
 export interface RouteBody<Body> {
   /** Schema for the validated provider-native body sent as the JSON request. */
@@ -58,6 +56,7 @@ export interface Route<Body, Prepared = unknown> {
     prepared: Prepared,
     request: LLMRequest,
     runtime: TransportRuntime,
+    options?: StreamOptions,
   ) => Stream.Stream<LLMEvent, AIError>
 }
 
@@ -157,6 +156,7 @@ export interface Interface {
 
 export interface StreamOptions {
   readonly http?: HttpMiddleware
+  readonly webSocket?: WebSocketChannelExecutor
 }
 
 export interface StreamMethod {
@@ -314,23 +314,29 @@ function makeFromTransport<Body, Prepared, Frame, Event, State>(
           encodeBody,
           headers: routeInput.headers,
           middleware: options?.http,
+          webSocket: options?.webSocket,
         }),
-      streamPrepared: (prepared: Prepared, request: LLMRequest, runtime: TransportRuntime) => {
+      streamPrepared: (prepared: Prepared, request: LLMRequest, runtime: TransportRuntime, options?: StreamOptions) => {
         const route = `${request.model.provider}/${request.model.route.id}`
-        const events = routeInput.transport
-          .frames(prepared, request, runtime)
-          .pipe(
-            Stream.mapEffect(decodeEvent(route)),
-            protocol.stream.terminal ? Stream.takeUntil(protocol.stream.terminal) : (stream) => stream,
-          )
-        return events.pipe(
-          Stream.mapAccumEffect(
-            () => protocol.stream.initial(request),
-            protocol.stream.step,
-            protocol.stream.onHalt ? { onHalt: protocol.stream.onHalt } : undefined,
+        return Stream.unwrap(
+          routeInput.transport.execute(prepared, request, runtime, options).pipe(
+            Effect.map((execution) => {
+              const events = execution.frames.pipe(
+                Stream.mapEffect(decodeEvent(route)),
+                protocol.stream.terminal ? Stream.takeUntil(protocol.stream.terminal) : (stream) => stream,
+              )
+              const stream = events.pipe(
+                Stream.mapAccumEffect(
+                  () => protocol.stream.initial(request),
+                  protocol.stream.step,
+                  protocol.stream.onHalt ? { onHalt: protocol.stream.onHalt } : undefined,
+                ),
+                Stream.catchCause((cause) => Stream.fail(streamError(route, `Failed to read ${route} stream`, cause))),
+                requireTerminalEvent(route),
+              )
+              return execution.complete ? stream.pipe(Stream.onEnd(execution.complete)) : stream
+            }),
           ),
-          Stream.catchCause((cause) => Stream.fail(streamError(route, `Failed to read ${route} stream`, cause))),
-          requireTerminalEvent(route),
         )
       },
     } satisfies Route<Body, Prepared>
@@ -413,7 +419,7 @@ const streamRequestWith = (runtime: TransportRuntime) => (request: LLMRequest, o
   Stream.unwrap(
     Effect.gen(function* () {
       const compiled = yield* compile(request, options)
-      return compiled.route.streamPrepared(compiled.prepared, compiled.request, runtime)
+      return compiled.route.streamPrepared(compiled.prepared, compiled.request, runtime, options)
     }),
   )
 
@@ -451,7 +457,6 @@ export const layer: Layer.Layer<Service, never, RequestExecutor.Service> = Layer
   Effect.gen(function* () {
     const stream = streamRequestWith({
       http: yield* RequestExecutor.Service,
-      webSocket: Option.getOrUndefined(yield* Effect.serviceOption(WebSocketExecutor.Service)),
     })
     return Service.of({ stream, generate: generateWith(stream) })
   }),

packages/ai/src/route/endpoint.ts

@@ -1,5 +1,5 @@
-import type { LLMRequest } from "../schema/index.js"
-import * as ProviderShared from "../protocols/shared.js"
+import type { LLMRequest } from "../schema"
+import * as ProviderShared from "../protocols/shared"
 
 export interface EndpointInput<Body> {
   readonly request: LLMRequest
@@ -53,4 +53,4 @@ export const render = <Body>(endpoint: Definition<Body>, input: EndpointInput<Bo
   return url
 }
 
-export * as Endpoint from "./endpoint.js"
+export * as Endpoint from "./endpoint"

packages/ai/src/route/executor.ts

@@ -1,4 +1,4 @@
-import { Cause, Context, Effect, Layer, Option, Schema, Stream } from "effect"
+import { Cause, Context, Effect, Layer } from "effect"
 import {
   FetchHttpClient,
   Headers,
@@ -14,8 +14,8 @@ import {
   HttpResponseDetails,
   AIError,
   TransportReason,
-} from "../schema/index.js"
-import { classifyProviderFailure } from "../provider-error.js"
+} from "../schema"
+import { classifyProviderFailure } from "../provider-error"
 
 export interface Interface {
   readonly execute: (
@@ -185,33 +185,21 @@ const secretValues = (request: HttpClientRequest.HttpClientRequest) => {
 // Two passes: structural (redact `"name": "value"` and `name=value` patterns
 // for any field name that looks sensitive) plus literal (replace any actual
 // secret values we sent in the request, in case the response echoes one back).
-const redactBody = (body: string, secrets: ReadonlySet<string>) =>
-  Array.from(secrets).reduce(
+const redactBody = (body: string, request: HttpClientRequest.HttpClientRequest) =>
+  Array.from(secretValues(request)).reduce(
     (text, secret) => text.split(secret).join(REDACTED),
     body.replace(REDACT_JSON_FIELD, `$1"${REDACTED}"`).replace(REDACT_QUERY_FIELD, `$1${REDACTED}`),
   )
 
-const responseBody = (body: string | void, secrets: ReadonlySet<string>) => {
+const responseBody = (body: string | void, request: HttpClientRequest.HttpClientRequest) => {
   if (body === undefined) return {}
-  const redacted = redactBody(body, secrets)
+  const redacted = redactBody(body, request)
   if (redacted.length <= BODY_LIMIT) return { body: redacted }
   return { body: redacted.slice(0, BODY_LIMIT), bodyTruncated: true }
 }
 
-const decodeProviderBody = Schema.decodeUnknownOption(
-  Schema.fromJsonString(
-    Schema.Struct({
-      message: Schema.optionalKey(Schema.String),
-      error: Schema.optionalKey(Schema.Struct({ message: Schema.optionalKey(Schema.String) })),
-    }),
-  ),
-)
-
 const providerMessage = (status: number, body: { readonly body?: string }) => {
-  if (body.body && body.body.length <= 500) {
-    const decoded = Option.getOrUndefined(decodeProviderBody(body.body))
-    return `Provider request failed with HTTP ${status}: ${decoded?.error?.message ?? decoded?.message ?? body.body}`
-  }
+  if (body.body && body.body.length <= 500) return `Provider request failed with HTTP ${status}: ${body.body}`
   return `Provider request failed with HTTP ${status}`
 }
 
@@ -240,7 +228,7 @@ const statusError =
       const headers = normalizedHeaders(response.headers)
       const retryAfter = retryAfterMs(headers)
       const rateLimit = rateLimitDetails(headers, retryAfter)
-      const details = responseBody(body, secretValues(request))
+      const details = responseBody(body, request)
       return yield* new AIError({
         module: "RequestExecutor",
         method: "execute",
@@ -261,122 +249,44 @@ const statusError =
       })
     })
 
-// Classifies an HTTP failure captured outside the executor (for example by the
-// AI SDK's own fetch) onto the same reason types and redacted HttpContext that
-// executor-driven requests produce. The originating request is not available on
-// that path, so the method is assumed (language model calls are always POST),
-// request headers are empty, and only structural body redaction applies.
-export const classifyHttpFailure = (input: {
-  readonly message: string
-  readonly url: string
-  readonly status?: number | undefined
-  readonly code?: string | undefined
-  readonly responseHeaders?: Record<string, string> | undefined
-  readonly responseBody?: string | undefined
-}) => {
-  const headers = normalizedHeaders(Headers.fromInput(input.responseHeaders))
-  const retryAfter = retryAfterMs(headers)
-  const rateLimit = rateLimitDetails(headers, retryAfter)
-  const details = responseBody(input.responseBody ?? undefined, new Set<string>())
-  return classifyProviderFailure({
-    message: input.message,
-    status: input.status,
-    code: input.code,
-    retryAfterMs: retryAfter,
-    rateLimit,
-    http: new HttpContext({
-      request: new HttpRequestDetails({ method: "POST", url: redactUrl(input.url), headers: {} }),
-      response:
-        input.status === undefined
-          ? undefined
-          : new HttpResponseDetails({ status: input.status, headers: redactHeaders(Headers.fromInput(headers), []) }),
-      ...details,
-      requestId: requestId(headers),
-      rateLimit,
-    }),
-  })
-}
-
-type HttpOperation = "request" | "read"
-
-const NativeTransportFailure = Schema.Struct({
-  message: Schema.String,
-  code: Schema.optionalKey(Schema.String),
-  cause: Schema.optionalKey(Schema.Unknown),
-})
-const decodeNativeTransportFailure = Schema.decodeUnknownOption(NativeTransportFailure)
-
-const nativeTransportFailure = (error: unknown) => {
-  const failure = Option.getOrUndefined(decodeNativeTransportFailure(error))
-  if (!failure) return undefined
-  if (failure.code !== undefined) return failure
-  const cause = Option.getOrUndefined(decodeNativeTransportFailure(failure.cause))
-  if (cause?.code !== undefined) return cause
-  return failure
-}
-
-const httpError = (input: {
-  readonly error: unknown
-  readonly request: HttpClientRequest.HttpClientRequest
-  readonly operation: HttpOperation
-  readonly redactedNames: ReadonlyArray<string | RegExp>
-}) => {
-  const request = HttpClientError.isHttpClientError(input.error) ? input.error.request : input.request
-  const transportError = (failure: { readonly message: string; readonly code?: string | undefined }) =>
+const toHttpError = (redactedNames: ReadonlyArray<string | RegExp>) => (error: unknown) => {
+  const transportError = (input: {
+    readonly message: string
+    readonly kind?: string | undefined
+    readonly request?: HttpClientRequest.HttpClientRequest | undefined
+  }) =>
     new AIError({
       module: "RequestExecutor",
-      method: input.operation,
+      method: "execute",
       reason: new TransportReason({
-        message: failure.message,
-        transport: "http",
-        operation: input.operation,
-        code: failure.code,
-        url: redactUrl(request.url),
-        http: new HttpContext({ request: requestDetails(request, input.redactedNames) }),
+        message: input.message,
+        kind: input.kind,
+        url: input.request ? redactUrl(input.request.url) : undefined,
+        http: input.request ? new HttpContext({ request: requestDetails(input.request, redactedNames) }) : undefined,
       }),
     })
 
-  const source =
-    HttpClientError.isHttpClientError(input.error) && "cause" in input.error.reason
-      ? input.error.reason.cause
-      : input.error
-  const native = nativeTransportFailure(source)
-  const code = native?.code
-  const raw = native?.message ?? (input.error instanceof Error ? input.error.message : undefined)
-  const detail = raw ? redactBody(raw, secretValues(request)) : undefined
-  const message = code && detail && !detail.includes(code) ? `${code}: ${detail}` : detail
-
-  if (Cause.isTimeoutError(input.error) || Cause.isTimeoutError(source))
-    return transportError({ message: message ?? "HTTP transport timed out", code: code ?? "Timeout" })
-  if (!HttpClientError.isHttpClientError(input.error))
-    return transportError({ message: message ?? "HTTP transport failed", code })
-  if (input.error.reason._tag === "TransportError") {
+  if (Cause.isTimeoutError(error)) {
+    return transportError({ message: error.message, kind: "Timeout" })
+  }
+  if (!HttpClientError.isHttpClientError(error)) {
+    return transportError({ message: error instanceof Error ? error.message : "HTTP transport failed" })
+  }
+  const request = "request" in error ? error.request : undefined
+  if (error.reason._tag === "TransportError") {
     return transportError({
-      message: message ?? input.error.reason.description ?? "HTTP transport failed",
-      code: code ?? input.error.reason._tag,
+      message: error.reason.description ?? "HTTP transport failed",
+      kind: error.reason._tag,
+      request,
     })
   }
   return transportError({
-    message: message ?? `HTTP transport failed: ${input.error.reason._tag}`,
-    code: code ?? input.error.reason._tag,
+    message: `HTTP transport failed: ${error.reason._tag}`,
+    kind: error.reason._tag,
+    request,
   })
 }
 
-export const stream = (
-  executor: Interface,
-  request: HttpClientRequest.HttpClientRequest,
-  middleware?: HttpMiddleware,
-): Stream.Stream<Uint8Array, AIError> =>
-  Stream.unwrap(
-    Effect.gen(function* () {
-      const redactedNames = yield* Headers.CurrentRedactedNames
-      const response = yield* executor.execute(request, middleware)
-      return response.stream.pipe(
-        Stream.mapError((error) => httpError({ error, request: response.request, operation: "read", redactedNames })),
-      )
-    }),
-  )
-
 export const layer: Layer.Layer<Service, never, HttpClient.HttpClient> = Layer.effect(
   Service,
   Effect.gen(function* () {
@@ -385,16 +295,15 @@ export const layer: Layer.Layer<Service, never, HttpClient.HttpClient> = Layer.e
       Effect.gen(function* () {
         const redactedNames = yield* Headers.CurrentRedactedNames
         if (!middleware)
-          return yield* http.execute(request).pipe(
-            Effect.mapError((error) => httpError({ error, request, operation: "request", redactedNames })),
-            Effect.flatMap(statusError(request, redactedNames)),
-          )
+          return yield* http
+            .execute(request)
+            .pipe(Effect.mapError(toHttpError(redactedNames)), Effect.flatMap(statusError(request, redactedNames)))
 
         const response = yield* middleware(request, (input) =>
           http
             .execute(input)
             .pipe(Effect.mapError((cause) => (cause instanceof Error ? cause : new Error(String(cause))))),
-        ).pipe(Effect.mapError((error) => httpError({ error, request, operation: "request", redactedNames })))
+        ).pipe(Effect.mapError(toHttpError(redactedNames)))
         return yield* statusError(response.request, redactedNames)(response)
       })
     return Service.of({
@@ -405,4 +314,4 @@ export const layer: Layer.Layer<Service, never, HttpClient.HttpClient> = Layer.e
 
 export const fetchLayer = layer.pipe(Layer.provide(FetchHttpClient.layer))
 
-export * as RequestExecutor from "./executor.js"
+export * as RequestExecutor from "./executor"