fix(task): handle active background continuation

.github/ISSUE_TEMPLATE/bug-report.yml

@@ -1,5 +1,5 @@
 name: Bug report
-description: Report an issue that should be fixed (avoid pasting giant AI generated summaries or your issue may be closed/ignored)
+description: Report an issue that should be fixed
 body:
   - type: textarea
     id: description

.opencode/skills/improve-codebase-architecture/DEEPENING.md

@@ -0,0 +1,37 @@
+# Deepening
+
+How to deepen a cluster of shallow modules safely, given its dependencies. Assumes the vocabulary in [LANGUAGE.md](LANGUAGE.md) — **module**, **interface**, **seam**, **adapter**.
+
+## Dependency categories
+
+When assessing a candidate for deepening, classify its dependencies. The category determines how the deepened module is tested across its seam.
+
+### 1. In-process
+
+Pure computation, in-memory state, no I/O. Always deepenable — merge the modules and test through the new interface directly. No adapter needed.
+
+### 2. Local-substitutable
+
+Dependencies that have local test stand-ins (PGLite for Postgres, in-memory filesystem). Deepenable if the stand-in exists. The deepened module is tested with the stand-in running in the test suite. The seam is internal; no port at the module's external interface.
+
+### 3. Remote but owned (Ports & Adapters)
+
+Your own services across a network boundary (microservices, internal APIs). Define a **port** (interface) at the seam. The deep module owns the logic; the transport is injected as an **adapter**. Tests use an in-memory adapter. Production uses an HTTP/gRPC/queue adapter.
+
+Recommendation shape: _"Define a port at the seam, implement an HTTP adapter for production and an in-memory adapter for testing, so the logic sits in one deep module even though it's deployed across a network."_
+
+### 4. True external (Mock)
+
+Third-party services (Stripe, Twilio, etc.) you don't control. The deepened module takes the external dependency as an injected port; tests provide a mock adapter.
+
+## Seam discipline
+
+- **One adapter means a hypothetical seam. Two adapters means a real one.** Don't introduce a port unless at least two adapters are justified (typically production + test). A single-adapter seam is just indirection.
+- **Internal seams vs external seams.** A deep module can have internal seams (private to its implementation, used by its own tests) as well as the external seam at its interface. Don't expose internal seams through the interface just because tests use them.
+
+## Testing strategy: replace, don't layer
+
+- Old unit tests on shallow modules become waste once tests at the deepened module's interface exist — delete them.
+- Write new tests at the deepened module's interface. The **interface is the test surface**.
+- Tests assert on observable outcomes through the interface, not internal state.
+- Tests should survive internal refactors — they describe behaviour, not implementation. If a test has to change when the implementation changes, it's testing past the interface.

.opencode/skills/improve-codebase-architecture/INTERFACE-DESIGN.md

@@ -0,0 +1,44 @@
+# Interface Design
+
+When the user wants to explore alternative interfaces for a chosen deepening candidate, use this parallel sub-agent pattern. Based on "Design It Twice" (Ousterhout) — your first idea is unlikely to be the best.
+
+Uses the vocabulary in [LANGUAGE.md](LANGUAGE.md) — **module**, **interface**, **seam**, **adapter**, **leverage**.
+
+## Process
+
+### 1. Frame the problem space
+
+Before spawning sub-agents, write a user-facing explanation of the problem space for the chosen candidate:
+
+- The constraints any new interface would need to satisfy
+- The dependencies it would rely on, and which category they fall into (see [DEEPENING.md](DEEPENING.md))
+- A rough illustrative code sketch to ground the constraints — not a proposal, just a way to make the constraints concrete
+
+Show this to the user, then immediately proceed to Step 2. The user reads and thinks while the sub-agents work in parallel.
+
+### 2. Spawn sub-agents
+
+Spawn 3+ sub-agents in parallel using the Agent tool. Each must produce a **radically different** interface for the deepened module.
+
+Prompt each sub-agent with a separate technical brief (file paths, coupling details, dependency category from [DEEPENING.md](DEEPENING.md), what sits behind the seam). The brief is independent of the user-facing problem-space explanation in Step 1. Give each agent a different design constraint:
+
+- Agent 1: "Minimize the interface — aim for 1–3 entry points max. Maximise leverage per entry point."
+- Agent 2: "Maximise flexibility — support many use cases and extension."
+- Agent 3: "Optimise for the most common caller — make the default case trivial."
+- Agent 4 (if applicable): "Design around ports & adapters for cross-seam dependencies."
+
+Include both [LANGUAGE.md](LANGUAGE.md) vocabulary and CONTEXT.md vocabulary in the brief so each sub-agent names things consistently with the architecture language and the project's domain language.
+
+Each sub-agent outputs:
+
+1. Interface (types, methods, params — plus invariants, ordering, error modes)
+2. Usage example showing how callers use it
+3. What the implementation hides behind the seam
+4. Dependency strategy and adapters (see [DEEPENING.md](DEEPENING.md))
+5. Trade-offs — where leverage is high, where it's thin
+
+### 3. Present and compare
+
+Present designs sequentially so the user can absorb each one, then compare them in prose. Contrast by **depth** (leverage at the interface), **locality** (where change concentrates), and **seam placement**.
+
+After comparing, give your own recommendation: which design you think is strongest and why. If elements from different designs would combine well, propose a hybrid. Be opinionated — the user wants a strong read, not a menu.

.opencode/skills/improve-codebase-architecture/LANGUAGE.md

@@ -0,0 +1,53 @@
+# Language
+
+Shared vocabulary for every suggestion this skill makes. Use these terms exactly — don't substitute "component," "service," "API," or "boundary." Consistent language is the whole point.
+
+## Terms
+
+**Module**
+Anything with an interface and an implementation. Deliberately scale-agnostic — applies equally to a function, class, package, or tier-spanning slice.
+_Avoid_: unit, component, service.
+
+**Interface**
+Everything a caller must know to use the module correctly. Includes the type signature, but also invariants, ordering constraints, error modes, required configuration, and performance characteristics.
+_Avoid_: API, signature (too narrow — those refer only to the type-level surface).
+
+**Implementation**
+What's inside a module — its body of code. Distinct from **Adapter**: a thing can be a small adapter with a large implementation (a Postgres repo) or a large adapter with a small implementation (an in-memory fake). Reach for "adapter" when the seam is the topic; "implementation" otherwise.
+
+**Depth**
+Leverage at the interface — the amount of behaviour a caller (or test) can exercise per unit of interface they have to learn. A module is **deep** when a large amount of behaviour sits behind a small interface. A module is **shallow** when the interface is nearly as complex as the implementation.
+
+**Seam** _(from Michael Feathers)_
+A place where you can alter behaviour without editing in that place. The _location_ at which a module's interface lives. Choosing where to put the seam is its own design decision, distinct from what goes behind it.
+_Avoid_: boundary (overloaded with DDD's bounded context).
+
+**Adapter**
+A concrete thing that satisfies an interface at a seam. Describes _role_ (what slot it fills), not substance (what's inside).
+
+**Leverage**
+What callers get from depth. More capability per unit of interface they have to learn. One implementation pays back across N call sites and M tests.
+
+**Locality**
+What maintainers get from depth. Change, bugs, knowledge, and verification concentrate at one place rather than spreading across callers. Fix once, fixed everywhere.
+
+## Principles
+
+- **Depth is a property of the interface, not the implementation.** A deep module can be internally composed of small, mockable, swappable parts — they just aren't part of the interface. A module can have **internal seams** (private to its implementation, used by its own tests) as well as the **external seam** at its interface.
+- **The deletion test.** Imagine deleting the module. If complexity vanishes, the module wasn't hiding anything (it was a pass-through). If complexity reappears across N callers, the module was earning its keep.
+- **The interface is the test surface.** Callers and tests cross the same seam. If you want to test _past_ the interface, the module is probably the wrong shape.
+- **One adapter means a hypothetical seam. Two adapters means a real one.** Don't introduce a seam unless something actually varies across it.
+
+## Relationships
+
+- A **Module** has exactly one **Interface** (the surface it presents to callers and tests).
+- **Depth** is a property of a **Module**, measured against its **Interface**.
+- A **Seam** is where a **Module**'s **Interface** lives.
+- An **Adapter** sits at a **Seam** and satisfies the **Interface**.
+- **Depth** produces **Leverage** for callers and **Locality** for maintainers.
+
+## Rejected framings
+
+- **Depth as ratio of implementation-lines to interface-lines** (Ousterhout): rewards padding the implementation. We use depth-as-leverage instead.
+- **"Interface" as the TypeScript `interface` keyword or a class's public methods**: too narrow — interface here includes every fact a caller must know.
+- **"Boundary"**: overloaded with DDD's bounded context. Say **seam** or **interface**.

.opencode/skills/improve-codebase-architecture/SKILL.md

@@ -0,0 +1,71 @@
+---
+name: improve-codebase-architecture
+description: Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.
+---
+
+# Improve Codebase Architecture
+
+Surface architectural friction and propose **deepening opportunities** — refactors that turn shallow modules into deep ones. The aim is testability and AI-navigability.
+
+## Glossary
+
+Use these terms exactly in every suggestion. Consistent language is the point — don't drift into "component," "service," "API," or "boundary." Full definitions in [LANGUAGE.md](LANGUAGE.md).
+
+- **Module** — anything with an interface and an implementation (function, class, package, slice).
+- **Interface** — everything a caller must know to use the module: types, invariants, error modes, ordering, config. Not just the type signature.
+- **Implementation** — the code inside.
+- **Depth** — leverage at the interface: a lot of behaviour behind a small interface. **Deep** = high leverage. **Shallow** = interface nearly as complex as the implementation.
+- **Seam** — where an interface lives; a place behaviour can be altered without editing in place. (Use this, not "boundary.")
+- **Adapter** — a concrete thing satisfying an interface at a seam.
+- **Leverage** — what callers get from depth.
+- **Locality** — what maintainers get from depth: change, bugs, knowledge concentrated in one place.
+
+Key principles (see [LANGUAGE.md](LANGUAGE.md) for the full list):
+
+- **Deletion test**: imagine deleting the module. If complexity vanishes, it was a pass-through. If complexity reappears across N callers, it was earning its keep.
+- **The interface is the test surface.**
+- **One adapter = hypothetical seam. Two adapters = real seam.**
+
+This skill is _informed_ by the project's domain model. The domain language gives names to good seams; ADRs record decisions the skill should not re-litigate.
+
+## Process
+
+### 1. Explore
+
+Read the project's domain glossary and any ADRs in the area you're touching first.
+
+Then use the Agent tool with `subagent_type=Explore` to walk the codebase. Don't follow rigid heuristics — explore organically and note where you experience friction:
+
+- Where does understanding one concept require bouncing between many small modules?
+- Where are modules **shallow** — interface nearly as complex as the implementation?
+- Where have pure functions been extracted just for testability, but the real bugs hide in how they're called (no **locality**)?
+- Where do tightly-coupled modules leak across their seams?
+- Which parts of the codebase are untested, or hard to test through their current interface?
+
+Apply the **deletion test** to anything you suspect is shallow: would deleting it concentrate complexity, or just move it? A "yes, concentrates" is the signal you want.
+
+### 2. Present candidates
+
+Present a numbered list of deepening opportunities. For each candidate:
+
+- **Files** — which files/modules are involved
+- **Problem** — why the current architecture is causing friction
+- **Solution** — plain English description of what would change
+- **Benefits** — explained in terms of locality and leverage, and also in how tests would improve
+
+**Use CONTEXT.md vocabulary for the domain, and [LANGUAGE.md](LANGUAGE.md) vocabulary for the architecture.** If `CONTEXT.md` defines "Order," talk about "the Order intake module" — not "the FooBarHandler," and not "the Order service."
+
+**ADR conflicts**: if a candidate contradicts an existing ADR, only surface it when the friction is real enough to warrant revisiting the ADR. Mark it clearly (e.g. _"contradicts ADR-0007 — but worth reopening because…"_). Don't list every theoretical refactor an ADR forbids.
+
+Do NOT propose interfaces yet. Ask the user: "Which of these would you like to explore?"
+
+### 3. Grilling loop
+
+Once the user picks a candidate, drop into a grilling conversation. Walk the design tree with them — constraints, dependencies, the shape of the deepened module, what sits behind the seam, what tests survive.
+
+Side effects happen inline as decisions crystallize:
+
+- **Naming a deepened module after a concept not in `CONTEXT.md`?** Add the term to `CONTEXT.md` — same discipline as `/grill-with-docs` (see [CONTEXT-FORMAT.md](../grill-with-docs/CONTEXT-FORMAT.md)). Create the file lazily if it doesn't exist.
+- **Sharpening a fuzzy term during the conversation?** Update `CONTEXT.md` right there.
+- **User rejects the candidate with a load-bearing reason?** Offer an ADR, framed as: _"Want me to record this as an ADR so future architecture reviews don't re-suggest it?"_ Only offer when the reason would actually be needed by a future explorer to avoid re-suggesting the same thing — skip ephemeral reasons ("not worth it right now") and self-evident ones. See [ADR-FORMAT.md](../grill-with-docs/ADR-FORMAT.md).
+- **Want to explore alternative interfaces for the deepened module?** See [INTERFACE-DESIGN.md](INTERFACE-DESIGN.md).

bun.lock

@@ -29,7 +29,7 @@
     },
     "packages/app": {
       "name": "@opencode-ai/app",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@kobalte/core": "catalog:",
         "@opencode-ai/core": "workspace:*",
@@ -83,26 +83,9 @@
         "vite-plugin-solid": "catalog:",
       },
     },
-    "packages/cli": {
-      "name": "@opencode-ai/cli",
-      "version": "1.15.13",
-      "bin": {
-        "opencode": "./src/index.ts",
-      },
-      "dependencies": {
-        "@effect/platform-node": "catalog:",
-        "@opencode-ai/core": "workspace:*",
-        "effect": "catalog:",
-      },
-      "devDependencies": {
-        "@tsconfig/bun": "catalog:",
-        "@types/bun": "catalog:",
-        "@typescript/native-preview": "catalog:",
-      },
-    },
     "packages/console/app": {
       "name": "@opencode-ai/console-app",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@cloudflare/vite-plugin": "1.15.2",
         "@ibm/plex": "6.4.1",
@@ -138,7 +121,7 @@
     },
     "packages/console/core": {
       "name": "@opencode-ai/console-core",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@aws-sdk/client-sts": "3.782.0",
         "@jsx-email/render": "1.1.1",
@@ -165,7 +148,7 @@
     },
     "packages/console/function": {
       "name": "@opencode-ai/console-function",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@ai-sdk/anthropic": "3.0.64",
         "@ai-sdk/openai": "3.0.48",
@@ -187,7 +170,7 @@
     },
     "packages/console/mail": {
       "name": "@opencode-ai/console-mail",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@jsx-email/all": "2.2.3",
         "@jsx-email/cli": "1.4.3",
@@ -211,7 +194,7 @@
     },
     "packages/console/support": {
       "name": "@opencode-ai/console-support",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@cloudflare/vite-plugin": "1.15.2",
         "@opencode-ai/console-core": "workspace:*",
@@ -231,7 +214,7 @@
     },
     "packages/core": {
       "name": "@opencode-ai/core",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "bin": {
         "opencode": "./bin/opencode",
       },
@@ -269,11 +252,10 @@
         "ai-gateway-provider": "3.1.2",
         "cross-spawn": "catalog:",
         "effect": "catalog:",
-        "gitlab-ai-provider": "6.8.0",
+        "gitlab-ai-provider": "6.7.0",
         "glob": "13.0.5",
         "google-auth-library": "10.5.0",
         "immer": "11.1.4",
-        "jsonc-parser": "3.3.1",
         "mime-types": "3.0.2",
         "minimatch": "10.2.5",
         "npm-package-arg": "13.0.2",
@@ -293,7 +275,7 @@
     },
     "packages/desktop": {
       "name": "@opencode-ai/desktop",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@zip.js/zip.js": "2.7.62",
         "drizzle-orm": "catalog:",
@@ -330,12 +312,12 @@
         "zod-openapi": "5.4.6",
       },
       "optionalDependencies": {
-        "@lydell/node-pty-darwin-arm64": "1.2.0-beta.12",
-        "@lydell/node-pty-darwin-x64": "1.2.0-beta.12",
-        "@lydell/node-pty-linux-arm64": "1.2.0-beta.12",
-        "@lydell/node-pty-linux-x64": "1.2.0-beta.12",
-        "@lydell/node-pty-win32-arm64": "1.2.0-beta.12",
-        "@lydell/node-pty-win32-x64": "1.2.0-beta.12",
+        "@lydell/node-pty-darwin-arm64": "1.2.0-beta.10",
+        "@lydell/node-pty-darwin-x64": "1.2.0-beta.10",
+        "@lydell/node-pty-linux-arm64": "1.2.0-beta.10",
+        "@lydell/node-pty-linux-x64": "1.2.0-beta.10",
+        "@lydell/node-pty-win32-arm64": "1.2.0-beta.10",
+        "@lydell/node-pty-win32-x64": "1.2.0-beta.10",
         "@parcel/watcher-darwin-arm64": "2.5.1",
         "@parcel/watcher-darwin-x64": "2.5.1",
         "@parcel/watcher-linux-arm64-glibc": "2.5.1",
@@ -348,7 +330,7 @@
     },
     "packages/effect-drizzle-sqlite": {
       "name": "@opencode-ai/effect-drizzle-sqlite",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "drizzle-orm": "catalog:",
         "effect": "catalog:",
@@ -362,7 +344,7 @@
     },
     "packages/enterprise": {
       "name": "@opencode-ai/enterprise",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@opencode-ai/core": "workspace:*",
         "@opencode-ai/ui": "workspace:*",
@@ -392,7 +374,7 @@
     },
     "packages/function": {
       "name": "@opencode-ai/function",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@octokit/auth-app": "8.0.1",
         "@octokit/rest": "catalog:",
@@ -408,7 +390,7 @@
     },
     "packages/http-recorder": {
       "name": "@opencode-ai/http-recorder",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@effect/platform-node": "catalog:",
         "effect": "catalog:",
@@ -421,7 +403,7 @@
     },
     "packages/llm": {
       "name": "@opencode-ai/llm",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@smithy/eventstream-codec": "4.2.14",
         "@smithy/util-utf8": "4.2.2",
@@ -439,7 +421,7 @@
     },
     "packages/opencode": {
       "name": "opencode",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "bin": {
         "opencode": "./bin/opencode",
       },
@@ -510,7 +492,7 @@
         "drizzle-orm": "catalog:",
         "effect": "catalog:",
         "fuzzysort": "3.1.0",
-        "gitlab-ai-provider": "6.8.0",
+        "gitlab-ai-provider": "6.7.0",
         "glob": "13.0.5",
         "google-auth-library": "10.5.0",
         "gray-matter": "4.0.3",
@@ -579,7 +561,7 @@
     },
     "packages/plugin": {
       "name": "@opencode-ai/plugin",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@opencode-ai/sdk": "workspace:*",
         "effect": "catalog:",
@@ -617,7 +599,7 @@
     },
     "packages/sdk/js": {
       "name": "@opencode-ai/sdk",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "cross-spawn": "catalog:",
       },
@@ -632,7 +614,7 @@
     },
     "packages/slack": {
       "name": "@opencode-ai/slack",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@opencode-ai/sdk": "workspace:*",
         "@slack/bolt": "^3.17.1",
@@ -645,9 +627,8 @@
     },
     "packages/stats/app": {
       "name": "@opencode-ai/stats-app",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
-        "@ibm/plex": "6.4.1",
         "@opencode-ai/stats-core": "workspace:*",
         "@opencode-ai/ui": "workspace:*",
         "@solidjs/meta": "catalog:",
@@ -657,7 +638,6 @@
         "effect": "catalog:",
         "nitro": "3.0.1-alpha.1",
         "solid-js": "catalog:",
-        "sst": "catalog:",
         "vite": "catalog:",
       },
       "devDependencies": {
@@ -670,7 +650,7 @@
     },
     "packages/stats/core": {
       "name": "@opencode-ai/stats-core",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@aws-sdk/client-athena": "3.933.0",
         "@planetscale/database": "1.19.0",
@@ -689,7 +669,7 @@
     },
     "packages/stats/server": {
       "name": "@opencode-ai/stats-server",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@aws-sdk/client-firehose": "3.933.0",
         "@effect/platform-node": "catalog:",
@@ -729,7 +709,7 @@
     },
     "packages/ui": {
       "name": "@opencode-ai/ui",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@kobalte/core": "catalog:",
         "@opencode-ai/core": "workspace:*",
@@ -778,7 +758,7 @@
     },
     "packages/web": {
       "name": "@opencode-ai/web",
-      "version": "1.15.13",
+      "version": "1.15.11",
       "dependencies": {
         "@astrojs/cloudflare": "12.6.3",
         "@astrojs/markdown-remark": "6.3.1",
@@ -842,7 +822,7 @@
     "@effect/sql-sqlite-bun": "4.0.0-beta.66",
     "@hono/zod-validator": "0.4.2",
     "@kobalte/core": "0.13.11",
-    "@lydell/node-pty": "1.2.0-beta.12",
+    "@lydell/node-pty": "1.2.0-beta.10",
     "@npmcli/arborist": "9.4.0",
     "@octokit/rest": "22.0.0",
     "@openauthjs/openauth": "0.0.0-20250322224806",
@@ -1526,19 +1506,19 @@
 
     "@lukeed/ms": ["@lukeed/ms@2.0.2", "", {}, "sha512-9I2Zn6+NJLfaGoz9jN3lpwDgAYvfGeNYdbAIjJOqzs4Tpc+VU3Jqq4IofSUBKajiDS8k9fZIg18/z13mpk1bsA=="],
 
-    "@lydell/node-pty": ["@lydell/node-pty@1.2.0-beta.12", "", { "optionalDependencies": { "@lydell/node-pty-darwin-arm64": "1.2.0-beta.12", "@lydell/node-pty-darwin-x64": "1.2.0-beta.12", "@lydell/node-pty-linux-arm64": "1.2.0-beta.12", "@lydell/node-pty-linux-x64": "1.2.0-beta.12", "@lydell/node-pty-win32-arm64": "1.2.0-beta.12", "@lydell/node-pty-win32-x64": "1.2.0-beta.12" } }, "sha512-qIK890UwPupoj07osVvgOIa++1mxeHbcGry4PKRHhNVNs81V2SCG34eJr46GybiOmBtc8Sj5PB1/GGM5PL549g=="],
+    "@lydell/node-pty": ["@lydell/node-pty@1.2.0-beta.10", "", { "optionalDependencies": { "@lydell/node-pty-darwin-arm64": "1.2.0-beta.10", "@lydell/node-pty-darwin-x64": "1.2.0-beta.10", "@lydell/node-pty-linux-arm64": "1.2.0-beta.10", "@lydell/node-pty-linux-x64": "1.2.0-beta.10", "@lydell/node-pty-win32-arm64": "1.2.0-beta.10", "@lydell/node-pty-win32-x64": "1.2.0-beta.10" } }, "sha512-Fv+A3+MZVA8qhkBIZsM1E6dCdHNMyXXz22mAYiMWd03LlyK///F3OH6CKPX9mj4id7LUlxpr45yPzyBVy9aDPw=="],
 
-    "@lydell/node-pty-darwin-arm64": ["@lydell/node-pty-darwin-arm64@1.2.0-beta.12", "", { "os": "darwin", "cpu": "arm64" }, "sha512-tqaifcY9Cr41SblO1+FLzh8oxxtkNhuW9Dhl22lKme9BreYvKvxEZcdPIXTuqkJc5tagOEC4QHShKmJjLyLXLQ=="],
+    "@lydell/node-pty-darwin-arm64": ["@lydell/node-pty-darwin-arm64@1.2.0-beta.10", "", { "os": "darwin", "cpu": "arm64" }, "sha512-C+eqDyRNHRYvx7RaHj6VVCx6nCpRBPuuxhTcc3JH3GuBMoxTsYeY4GkWH2XOktrgbAq1BG8e/Y8bu/wNQreCEw=="],
 
-    "@lydell/node-pty-darwin-x64": ["@lydell/node-pty-darwin-x64@1.2.0-beta.12", "", { "os": "darwin", "cpu": "x64" }, "sha512-4LrS5pCJwqHKDVf1zS2gyNV0m4hKAXch+XZNhbZ6LY8uwVL8BhchzQBO40Os5anuRxRCWzHpw4Sp64Ie8q7E4Q=="],
+    "@lydell/node-pty-darwin-x64": ["@lydell/node-pty-darwin-x64@1.2.0-beta.10", "", { "os": "darwin", "cpu": "x64" }, "sha512-aZoIK6HtJO5BiT4ELm683U4dyHtt8b7wNgq3NJqYAQwSXrcPv576Z8vY3BIulVxfcFkht/SPLKou9TtdFXdNpg=="],
 
-    "@lydell/node-pty-linux-arm64": ["@lydell/node-pty-linux-arm64@1.2.0-beta.12", "", { "os": "linux", "cpu": "arm64" }, "sha512-Sx+A71x5BDGHt9ansfrtGxwq2VFVDWvJUAdlUL0Hv0qeiJUfts+hgopx+CgT4PSwahKjdEgtu0+FAfY9rICKRw=="],
+    "@lydell/node-pty-linux-arm64": ["@lydell/node-pty-linux-arm64@1.2.0-beta.10", "", { "os": "linux", "cpu": "arm64" }, "sha512-0cKX2iMyXFNBE4fGtGK6B7IkdXcDMZajyEDoGMOgQQs/DDtoI5tSPcBcqNY9VitVrsRQA8+gFt6eKYU9Ye/lUA=="],
 
-    "@lydell/node-pty-linux-x64": ["@lydell/node-pty-linux-x64@1.2.0-beta.12", "", { "os": "linux", "cpu": "x64" }, "sha512-bJzs94njofYhGg/UDqW1nj0dtvvu+2OvxMY+RlLS1T17VgcktKoIR6PuenTwE5HJ/D6StCPADmXcT0nNsCKmIQ=="],
+    "@lydell/node-pty-linux-x64": ["@lydell/node-pty-linux-x64@1.2.0-beta.10", "", { "os": "linux", "cpu": "x64" }, "sha512-J9HnxvSzEeMH748+Ul1VrmCLWMo7iCVJy9EGijRR62+YO/Yk5GaCydUTZ+KzlH0/X5aTrgt5cfiof4vx45tRRg=="],
 
-    "@lydell/node-pty-win32-arm64": ["@lydell/node-pty-win32-arm64@1.2.0-beta.12", "", { "os": "win32", "cpu": "arm64" }, "sha512-p7POgjVEiFaBC3/y+AKuV1FzePCsJ6HmZDv2XK+jBZSfwP8+uBAw181ZiKYN1YuRa/XpmBGaWezcI8hZkbW++g=="],
+    "@lydell/node-pty-win32-arm64": ["@lydell/node-pty-win32-arm64@1.2.0-beta.10", "", { "os": "win32", "cpu": "arm64" }, "sha512-PlDJpJX/pnKyy6OmADKzhf+INZDDnzTBGaI0LT4laVNc6NblZNqUSkCMjLFWbeakeuQp0VG37M49WQSN9FDfeA=="],
 
-    "@lydell/node-pty-win32-x64": ["@lydell/node-pty-win32-x64@1.2.0-beta.12", "", { "os": "win32", "cpu": "x64" }, "sha512-IDFa00g7qUDGUYgByrUBJtC+mOjYVt/8KYyWivCg5JjGOHbBUACUQZLl0jTWmnr+tld/UyTpX90a2PY6oTVtRw=="],
+    "@lydell/node-pty-win32-x64": ["@lydell/node-pty-win32-x64@1.2.0-beta.10", "", { "os": "win32", "cpu": "x64" }, "sha512-ExFgWrzyldNAMi45U9PLIOu+g/RatP+f0c/dZxaooifME6yLW32BoHveH26/TtoAjZyJrc2iL0u48pgnR1fzmg=="],
 
     "@malept/cross-spawn-promise": ["@malept/cross-spawn-promise@2.0.0", "", { "dependencies": { "cross-spawn": "^7.0.1" } }, "sha512-1DpKU0Z5ThltBwjNySMC14g0CkbyhCaz9FkhxqNsZI6uAPJXFS8cMXlBKo26FJ8ZuW6S9GCMcR9IO5k2X5/9Fg=="],
 
@@ -1664,8 +1644,6 @@
 
     "@opencode-ai/app": ["@opencode-ai/app@workspace:packages/app"],
 
-    "@opencode-ai/cli": ["@opencode-ai/cli@workspace:packages/cli"],
-
     "@opencode-ai/console-app": ["@opencode-ai/console-app@workspace:packages/console/app"],
 
     "@opencode-ai/console-core": ["@opencode-ai/console-core@workspace:packages/console/core"],
@@ -3440,7 +3418,7 @@
 
     "github-slugger": ["github-slugger@2.0.0", "", {}, "sha512-IaOQ9puYtjrkq7Y0Ygl9KDZnrf/aiUJYUpVf89y8kyaxbRG7Y1SrX/jaumrv81vc61+kiMempujsM3Yw7w5qcw=="],
 
-    "gitlab-ai-provider": ["gitlab-ai-provider@6.8.0", "", { "dependencies": { "@anthropic-ai/sdk": "^0.71.0", "@anycable/core": "^0.9.2", "graphql-request": "^6.1.0", "isomorphic-ws": "^5.0.0", "openai": "^6.16.0", "socket.io-client": "^4.8.1", "vscode-jsonrpc": "^8.2.1", "zod": "^3.25.76" }, "peerDependencies": { "@ai-sdk/provider": ">=3.0.0", "@ai-sdk/provider-utils": ">=4.0.0" } }, "sha512-KwHASXkHtDcgrzTXZVp9Dyx6t8m9nK0R2fCm47MWcxxQ1kOBt3f2LZugtu1kOby8i4Sbd+kvBSYM66PGkDclng=="],
+    "gitlab-ai-provider": ["gitlab-ai-provider@6.7.0", "", { "dependencies": { "@anthropic-ai/sdk": "^0.71.0", "@anycable/core": "^0.9.2", "graphql-request": "^6.1.0", "isomorphic-ws": "^5.0.0", "openai": "^6.16.0", "socket.io-client": "^4.8.1", "vscode-jsonrpc": "^8.2.1", "zod": "^3.25.76" }, "peerDependencies": { "@ai-sdk/provider": ">=3.0.0", "@ai-sdk/provider-utils": ">=4.0.0" } }, "sha512-J7apROAmDwDJcLPcbQ9YQiPhyFN1Vvt/f3ugsePaUnzh3sedfHCHGETgG0Mm0dJCB2rC7DYwmV/sgxLQcQ/sjw=="],
 
     "glob": ["glob@13.0.5", "", { "dependencies": { "minimatch": "^10.2.1", "minipass": "^7.1.2", "path-scurry": "^2.0.0" } }, "sha512-BzXxZg24Ibra1pbQ/zE7Kys4Ua1ks7Bn6pKLkVPZ9FZe4JQS6/Q7ef3LG1H+k7lUf5l4T3PLSyYyYJVYUvfgTw=="],
 
@@ -4976,7 +4954,7 @@
 
     "uncrypto": ["uncrypto@0.1.3", "", {}, "sha512-Ql87qFHB3s/De2ClA9e0gsnS6zXG27SkTiSJwjCc9MebbfapQfuPzumMIUMi38ezPZVNFcHI9sUIepeQfw8J8Q=="],
 
-    "undici": ["undici@8.1.0", "", {}, "sha512-E9MkTS4xXLnRPYqxH2e6Hr2/49e7WFDKczKcCaFH4VaZs2iNvHMqeIkyUAD9vM8kujy9TjVrRlQ5KkdEJxB2pw=="],
+    "undici": ["undici@7.25.0", "", {}, "sha512-xXnp4kTyor2Zq+J1FfPI6Eq3ew5h6Vl0F/8d9XU5zZQf1tX9s2Su1/3PiMmUANFULpmksxkClamIZcaUqryHsQ=="],
 
     "undici-types": ["undici-types@7.16.0", "", {}, "sha512-Zz+aZWSj8LE6zoxD+xrjh4VfkIG8Ya6LvYkZqtUQGJPZjYl53ypCaUwWqo7eI0x66KBGeRo+mlBEkMSeSZ38Nw=="],
 
@@ -5528,6 +5506,8 @@
 
     "@dot/log/chalk": ["chalk@4.1.2", "", { "dependencies": { "ansi-styles": "^4.1.0", "supports-color": "^7.1.0" } }, "sha512-oKnbhFyRIXpUuez8iBMmyEa4nbj4IOQyuhc/wy9kY7/WVPcwIO9VA668Pu8RkO7+0G76SLROeyw9CpQ061i4mA=="],
 
+    "@effect/platform-node/undici": ["undici@8.1.0", "", {}, "sha512-E9MkTS4xXLnRPYqxH2e6Hr2/49e7WFDKczKcCaFH4VaZs2iNvHMqeIkyUAD9vM8kujy9TjVrRlQ5KkdEJxB2pw=="],
+
     "@effect/platform-node-shared/ws": ["ws@8.20.0", "", { "peerDependencies": { "bufferutil": "^4.0.1", "utf-8-validate": ">=5.0.2" }, "optionalPeers": ["bufferutil", "utf-8-validate"] }, "sha512-sAt8BhgNbzCtgGbt2OxmpuryO63ZoDk/sqaB/znQm94T4fCEsy/yV+7CdC1kJhOU9lboAEU7R3kquuycDoibVA=="],
 
     "@electron/asar/commander": ["commander@5.1.0", "", {}, "sha512-P0CysNDQ7rtVw4QIQtm+MRxV66vKFSvlsQvGYXZWR3qFU0jlMKHZZZgw8e+8DSah4UDKMqnknRDQz+xuQXQ/Zg=="],
@@ -5662,8 +5642,6 @@
 
     "@opencode-ai/core/@ai-sdk/openai-compatible": ["@ai-sdk/openai-compatible@2.0.41", "", { "dependencies": { "@ai-sdk/provider": "3.0.8", "@ai-sdk/provider-utils": "4.0.23" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-kNAGINk71AlOXx10Dq/PXw4t/9XjdK8uxfpVElRwtSFMdeSiLVt58p9TPx4/FJD+hxZuVhvxYj9r42osxWq79g=="],
 
-    "@opencode-ai/core/semver": ["semver@7.8.1", "", { "bin": { "semver": "bin/semver.js" } }, "sha512-rkVq3IXh+4FDGch+KwzX3aV9W3kO54GyEgpvBzSyctDA6Xtd7RJQV1xmXbeQp5v7+VzLOfVqiutSE6GICgPFvg=="],
-
     "@opencode-ai/desktop/@actions/artifact": ["@actions/artifact@4.0.0", "", { "dependencies": { "@actions/core": "^1.10.0", "@actions/github": "^6.0.1", "@actions/http-client": "^2.1.0", "@azure/core-http": "^3.0.5", "@azure/storage-blob": "^12.15.0", "@octokit/core": "^5.2.1", "@octokit/plugin-request-log": "^1.0.4", "@octokit/plugin-retry": "^3.0.9", "@octokit/request": "^8.4.1", "@octokit/request-error": "^5.1.1", "@protobuf-ts/plugin": "^2.2.3-alpha.1", "archiver": "^7.0.1", "jwt-decode": "^3.1.2", "unzip-stream": "^0.3.1" } }, "sha512-HCc2jMJRAfviGFAh0FsOR/jNfWhirxl7W6z8zDtttt0GltwxBLdEIjLiweOPFl9WbyJRW1VWnPUSAixJqcWUMQ=="],
 
     "@opencode-ai/desktop/marked": ["marked@15.0.12", "", { "bin": { "marked": "bin/marked.js" } }, "sha512-8dD6FusOQSrpv9Z1rdNMdlSgQOIP880DHqnohobOmYLElGEqAL/JvxvuxZO16r4HtjTlfPRDC1hbvxC9dPN2nA=="],
@@ -5674,8 +5652,6 @@
 
     "@opencode-ai/llm/@smithy/util-utf8": ["@smithy/util-utf8@4.2.2", "", { "dependencies": { "@smithy/util-buffer-from": "^4.2.2", "tslib": "^2.6.2" } }, "sha512-75MeYpjdWRe8M5E3AW0O4Cx3UadweS+cwdXjwYGBW5h/gxxnbeZ877sLPX/ZJA9GVTlL/qG0dXP29JWFCD1Ayw=="],
 
-    "@opencode-ai/script/semver": ["semver@7.8.1", "", { "bin": { "semver": "bin/semver.js" } }, "sha512-rkVq3IXh+4FDGch+KwzX3aV9W3kO54GyEgpvBzSyctDA6Xtd7RJQV1xmXbeQp5v7+VzLOfVqiutSE6GICgPFvg=="],
-
     "@opencode-ai/ui/@solid-primitives/resize-observer": ["@solid-primitives/resize-observer@2.1.3", "", { "dependencies": { "@solid-primitives/event-listener": "^2.4.3", "@solid-primitives/rootless": "^1.5.2", "@solid-primitives/static-store": "^0.1.2", "@solid-primitives/utils": "^6.3.2" }, "peerDependencies": { "solid-js": "^1.6.12" } }, "sha512-zBLje5E06TgOg93S7rGPldmhDnouNGhvfZVKOp+oG2XU8snA+GoCSSCz1M+jpNAg5Ek2EakU5UVQqL152WmdXQ=="],
 
     "@opencode-ai/web/@shikijs/transformers": ["@shikijs/transformers@3.20.0", "", { "dependencies": { "@shikijs/core": "3.20.0", "@shikijs/types": "3.20.0" } }, "sha512-PrHHMRr3Q5W1qB/42kJW6laqFyWdhrPF2hNR9qjOm1xcSiAO3hAHo7HaVyHE6pMyevmy3i51O8kuGGXC78uK3g=="],
@@ -6002,8 +5978,6 @@
 
     "nitro/h3": ["h3@2.0.1-rc.5", "", { "dependencies": { "rou3": "^0.7.9", "srvx": "^0.9.1" }, "peerDependencies": { "crossws": "^0.4.1" }, "optionalPeers": ["crossws"] }, "sha512-qkohAzCab0nLzXNm78tBjZDvtKMTmtygS8BJLT3VPczAQofdqlFXDPkXdLMJN4r05+xqneG8snZJ0HgkERCZTg=="],
 
-    "nitro/undici": ["undici@7.25.0", "", {}, "sha512-xXnp4kTyor2Zq+J1FfPI6Eq3ew5h6Vl0F/8d9XU5zZQf1tX9s2Su1/3PiMmUANFULpmksxkClamIZcaUqryHsQ=="],
-
     "node-gyp-build-optional-packages/detect-libc": ["detect-libc@2.1.2", "", {}, "sha512-Btj2BOOO83o3WyH59e8MgXsxEQVcarkUOpEYrubB0urwnN10yQ364rsiByU11nZlqWYZm05i/of7io4mzihBtQ=="],
 
     "npm-run-path/path-key": ["path-key@4.0.0", "", {}, "sha512-haREypq7xkM7ErfgIyA0z+Bj4AGKlMSdlQE2jvJo6huWD1EdkKYV+G/T4nq0YEF2vgTT8kqMFKo1uHn950r4SQ=="],
@@ -6022,8 +5996,6 @@
 
     "opencode/minimatch": ["minimatch@10.0.3", "", { "dependencies": { "@isaacs/brace-expansion": "^5.0.0" } }, "sha512-IPZ167aShDZZUMdRk66cyQAW3qr0WzbHkPdMYa8bzZhlHhO3jALbKdxcaak7W9FfT2rZNpQuUu4Od7ILEpXSaw=="],
 
-    "opencode/semver": ["semver@7.8.1", "", { "bin": { "semver": "bin/semver.js" } }, "sha512-rkVq3IXh+4FDGch+KwzX3aV9W3kO54GyEgpvBzSyctDA6Xtd7RJQV1xmXbeQp5v7+VzLOfVqiutSE6GICgPFvg=="],
-
     "opencode-gitlab-auth/open": ["open@10.2.0", "", { "dependencies": { "default-browser": "^5.2.1", "define-lazy-prop": "^3.0.0", "is-inside-container": "^1.0.0", "wsl-utils": "^0.1.0" } }, "sha512-YgBpdJHPyQ2UE5x+hlSXcnejzAvD0b22U2OuAP+8OnlJT+PjWPxtgmGqKKc+RgTM63U9gN0YzrYc71R2WT/hTA=="],
 
     "opencode-poe-auth/open": ["open@10.2.0", "", { "dependencies": { "default-browser": "^5.2.1", "define-lazy-prop": "^3.0.0", "is-inside-container": "^1.0.0", "wsl-utils": "^0.1.0" } }, "sha512-YgBpdJHPyQ2UE5x+hlSXcnejzAvD0b22U2OuAP+8OnlJT+PjWPxtgmGqKKc+RgTM63U9gN0YzrYc71R2WT/hTA=="],

bunfig.toml

@@ -2,7 +2,7 @@
 exact = true
 # Only install newly resolved package versions published at least 3 days ago.
 minimumReleaseAge = 259200
-minimumReleaseAgeExcludes = ["@opentui/core", "@opentui/core-darwin-arm64", "@opentui/core-darwin-x64", "@opentui/core-linux-arm64", "@opentui/core-linux-x64", "@opentui/core-win32-arm64", "@opentui/core-win32-x64", "@opentui/keymap", "@opentui/solid", "gitlab-ai-provider"]
+minimumReleaseAgeExcludes = ["@opentui/core", "@opentui/core-darwin-arm64", "@opentui/core-darwin-x64", "@opentui/core-linux-arm64", "@opentui/core-linux-x64", "@opentui/core-win32-arm64", "@opentui/core-win32-x64", "@opentui/keymap", "@opentui/solid"]
 
 [test]
 root = "./do-not-run-tests-from-root"

infra/stats.ts

@@ -1,5 +1,4 @@
 import { lakeAthenaWorkgroup, lakeCatalog, lakeCluster, lakeQueryPermissions, lakeRegion, tableBucket } from "./lake"
-import { EMAILOCTOPUS_API_KEY } from "./app"
 
 const domain = (() => {
   if ($app.stage === "production") return "stats.opencode.ai"
@@ -164,15 +163,15 @@ new sst.x.DevCommand("StatsStudio", {
 // APP
 ////////////////
 
-export const app = new sst.cloudflare.x.SolidStart("Stats", {
-  path: "packages/stats/app",
-  buildCommand: "bun run build",
-  domain,
-  link: [database, EMAILOCTOPUS_API_KEY],
-  environment: {
-    PUBLIC_URL: `https://${domain}/stats`,
-  },
-})
+// export const app = new sst.cloudflare.x.SolidStart("Stats", {
+//   path: "packages/stats/app",
+//   buildCommand: "bun run build",
+//   domain,
+//   link: [database],
+//   environment: {
+//     PUBLIC_URL: `https://${domain}`,
+//   },
+// })
 
 ////////////////
 // SERVICES

nix/hashes.json

@@ -1,8 +1,8 @@
 {
   "nodeModules": {
-    "x86_64-linux": "sha256-aTweGlyK8w+A9X7FpZnbVh+czXAVlttnn7Efhzm+8kY=",
-    "aarch64-linux": "sha256-Gl+fwiGv/BC9u8xV4h0NIYbEac+LJSODLSTipWwdWII=",
-    "aarch64-darwin": "sha256-Vt6eVf9gb+A5nULpMgtrg2YNIQy3L//wNVThVkVbgyc=",
-    "x86_64-darwin": "sha256-0uHk7YnGTeIOLxajdj+CcAokfAIdDLFeS0VwN0mXRrc="
+    "x86_64-linux": "sha256-KFVkxs0jtN1EXvwuxFMOlWg6BDr8L6K/wJZ/N73p8cI=",
+    "aarch64-linux": "sha256-7OLglRBBt5prwb9e0I0iMTXi3MrDfDI+8xcDrxtq/JI=",
+    "aarch64-darwin": "sha256-Q7iE6xJyomDnvqUgsHh+FlCnJFI6vK0zJSfI9BBYitQ=",
+    "x86_64-darwin": "sha256-9lVsq2Bw74Vgphg6GSo/HxCtDXUyabxcS/vE1Um/Hfw="
   }
 }

package.json

@@ -10,7 +10,7 @@
     "dev:desktop": "bun --cwd packages/desktop dev",
     "dev:web": "bun --cwd packages/app dev",
     "dev:console": "ulimit -n 10240 2>/dev/null; bun run --cwd packages/console/app dev",
-    "dev:stats": "bun sst shell --stage=production -- bun run --cwd packages/stats/app dev",
+    "dev:stats": "bun sst shell --stage=dev -- bun run --cwd packages/stats/app dev",
     "dev:storybook": "bun --cwd packages/storybook storybook",
     "lint": "oxlint",
     "typecheck": "bun turbo typecheck",
@@ -87,7 +87,7 @@
       "@sentry/vite-plugin": "4.6.0",
       "solid-js": "1.9.10",
       "vite-plugin-solid": "2.11.10",
-      "@lydell/node-pty": "1.2.0-beta.12"
+      "@lydell/node-pty": "1.2.0-beta.10"
     }
   },
   "devDependencies": {

packages/app/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@opencode-ai/app",
-  "version": "1.15.13",
+  "version": "1.15.11",
   "description": "",
   "type": "module",
   "exports": {

packages/app/src/context/directory-sync.ts

@@ -178,7 +178,7 @@ export const createDirSyncContext = (directory: string, serverSync: ReturnType<t
   type Child = ReturnType<(typeof serverSync)["child"]>
   type Setter = Child[1]
 
-  const current = createMemo(() => serverSync.child(directory, { mcp: true }))
+  const current = createMemo(() => serverSync.child(directory))
   const target = (directory?: string) => {
     if (!directory || directory === directory) return current()
     return serverSync.child(directory)

packages/app/src/context/global-sync/bootstrap.test.ts

@@ -10,7 +10,6 @@ const provider = { all: new Map(), connected: [], default: {} } satisfies Normal
 
 describe("bootstrapDirectory", () => {
   test("marks a loading directory partial during bootstrap and complete after success", async () => {
-    const mcpReads: string[] = []
     const [store, setStore] = createStore<State>({
       status: "loading",
       agent: [],
@@ -45,7 +44,6 @@ describe("bootstrapDirectory", () => {
 
     await bootstrapDirectory({
       directory: "/project",
-      mcp: false,
       global: {
         config: {} satisfies Config,
         path: { state: "", config: "", worktree: "/project", directory: "/project", home: "/home" },
@@ -57,20 +55,10 @@ describe("bootstrapDirectory", () => {
         config: { get: async () => ({ data: {} }) },
         session: { status: async () => ({ data: {} }) },
         vcs: { get: async () => ({ data: undefined }) },
-        command: {
-          list: async () => {
-            mcpReads.push("command")
-            return { data: [] }
-          },
-        },
+        command: { list: async () => ({ data: [] }) },
         permission: { list: async () => ({ data: [] }) },
         question: { list: async () => ({ data: [] }) },
-        mcp: {
-          status: async () => {
-            mcpReads.push("status")
-            return { data: {} }
-          },
-        },
+        mcp: { status: async () => ({ data: {} }) },
         provider: { list: async () => ({ data: { all: [], connected: [], default: {} } }) },
       } as unknown as OpencodeClient,
       store,
@@ -86,6 +74,5 @@ describe("bootstrapDirectory", () => {
     await new Promise((resolve) => setTimeout(resolve, 80))
 
     expect(store.status).toBe("complete")
-    expect(mcpReads).toEqual([])
   })
 })

packages/app/src/context/global-sync/bootstrap.ts

@@ -198,7 +198,6 @@ export const loadPathQuery = (directory: string | null, sdk: OpencodeClient) =>
 
 export async function bootstrapDirectory(input: {
   directory: string
-  mcp: boolean
   sdk: OpencodeClient
   store: Store<State>
   setStore: SetStoreFunction<State>
@@ -251,7 +250,7 @@ export async function bootstrapDirectory(input: {
             if (next) input.vcsCache.setStore("value", next)
           }),
         ),
-      input.mcp && (() => retry(() => input.sdk.command.list().then((x) => input.setStore("command", x.data ?? [])))),
+      () => retry(() => input.sdk.command.list().then((x) => input.setStore("command", x.data ?? []))),
       () =>
         retry(() =>
           input.sdk.permission.list().then((x) => {
@@ -305,7 +304,7 @@ export async function bootstrapDirectory(input: {
           }),
         ),
       () => Promise.resolve(input.loadSessions(input.directory)),
-      input.mcp && (() => input.queryClient.fetchQuery(loadMcpQuery(input.directory, input.sdk))),
+      () => input.queryClient.fetchQuery(loadMcpQuery(input.directory, input.sdk)),
       () =>
         input.queryClient.fetchQuery(loadProvidersQuery(input.directory, input.sdk)).catch((err) => {
           const project = getFilename(input.directory)

packages/app/src/context/global-sync/child-store.test.ts

@@ -6,7 +6,6 @@ import type { State } from "./types"
 import type { QueryOptionsApi } from "../server-sync"
 
 let createChildStoreManager: typeof import("./child-store").createChildStoreManager
-const queryGroups: Array<() => { queries: Array<{ enabled?: boolean }> }> = []
 
 const child = () => createStore({} as State)
 const provider = { all: new Map(), connected: [], default: {} } satisfies NormalizedProviderListResponse
@@ -49,15 +48,12 @@ beforeAll(async () => {
     persisted: (_target: string, store: unknown[]) => [store[0], store[1], null, () => true],
   }))
   mock.module("@tanstack/solid-query", () => ({
-    useQueries: (options: () => { queries: Array<{ enabled?: boolean }> }) => {
-      queryGroups.push(options)
-      return [
-        { isLoading: false, data: { state: "", config: "", worktree: "", directory: "", home: "" } },
-        { isLoading: false, data: {} },
-        { isLoading: false, data: [] },
-        { isLoading: false, data: provider },
-      ]
-    },
+    useQueries: () => [
+      { isLoading: false, data: { state: "", config: "", worktree: "", directory: "", home: "" } },
+      { isLoading: false, data: {} },
+      { isLoading: false, data: [] },
+      { isLoading: false, data: provider },
+    ],
   }))
 
   createChildStoreManager = (await import("./child-store")).createChildStoreManager
@@ -77,7 +73,6 @@ describe("createChildStoreManager", () => {
       isBooting: () => false,
       isLoadingSessions: () => false,
       onBootstrap() {},
-      onMcp() {},
       onDispose() {},
       translate: (key) => key,
       queryOptions: queryOptionsApi,
@@ -108,7 +103,6 @@ describe("createChildStoreManager", () => {
         onBootstrap(directory) {
           bootstraps.push(directory)
         },
-        onMcp() {},
         onDispose() {},
         translate: (key) => key,
         queryOptions: queryOptionsApi,
@@ -127,45 +121,4 @@ describe("createChildStoreManager", () => {
       dispose()
     }
   })
-
-  test("enables MCP only when requested for the directory", () => {
-    let manager: ReturnType<typeof createChildStoreManager> | undefined
-    const offset = queryGroups.length
-    const mcpLoads: string[] = []
-
-    const dispose = createOwner((owner) => {
-      manager = createChildStoreManager({
-        owner,
-        isBooting: () => false,
-        isLoadingSessions: () => false,
-        onBootstrap() {},
-        onMcp(directory) {
-          mcpLoads.push(directory)
-        },
-        onDispose() {},
-        translate: (key) => key,
-        queryOptions: queryOptionsApi,
-        global: { provider },
-      })
-    })
-
-    try {
-      if (!manager) throw new Error("manager required")
-      const [, setStore] = manager.child("/project", { bootstrap: false })
-      const queries = queryGroups[offset]
-      if (!queries) throw new Error("queries required")
-      expect(queries().queries[1]?.enabled).toBe(false)
-
-      setStore("status", "complete")
-      manager.child("/project", { bootstrap: false, mcp: true })
-      expect(queries().queries[1]?.enabled).toBe(true)
-      expect(mcpLoads).toEqual(["/project"])
-
-      manager.disableMcp("/project")
-      expect(queries().queries[1]?.enabled).toBe(false)
-      expect(manager.mcp("/project")).toBe(false)
-    } finally {
-      dispose()
-    }
-  })
 })

packages/app/src/context/global-sync/child-store.ts

@@ -1,4 +1,4 @@
-import { createRoot, createSignal, getOwner, onCleanup, runWithOwner, type Owner } from "solid-js"
+import { createRoot, getOwner, onCleanup, runWithOwner, type Owner } from "solid-js"
 import { createStore, type SetStoreFunction, type Store } from "solid-js/store"
 import { Persist, persisted } from "@/utils/persist"
 import type { VcsInfo } from "@opencode-ai/sdk/v2/client"
@@ -24,7 +24,6 @@ export function createChildStoreManager(input: {
   isBooting: (directory: string) => boolean
   isLoadingSessions: (directory: string) => boolean
   onBootstrap: (directory: string) => void
-  onMcp: (directory: string, setStore: SetStoreFunction<State>) => void
   onDispose: (directory: string) => void
   translate: (key: string, vars?: Record<string, string | number>) => string
   queryOptions: QueryOptionsApi
@@ -40,8 +39,6 @@ export function createChildStoreManager(input: {
   const pins = new Map<string, number>()
   const ownerPins = new WeakMap<object, Set<string>>()
   const disposers = new Map<string, () => void>()
-  const mcpDirectories = new Set<string>()
-  const mcpToggles = new Map<string, (enabled: boolean) => void>()
 
   const markKey = (key: DirectoryKey) => {
     if (!key) return
@@ -113,8 +110,6 @@ export function createChildStoreManager(input: {
     metaCache.delete(key)
     iconCache.delete(key)
     lifecycle.delete(key)
-    mcpDirectories.delete(key)
-    mcpToggles.delete(key)
     const dispose = disposers.get(key)
     if (dispose) {
       dispose()
@@ -178,12 +173,11 @@ export function createChildStoreManager(input: {
         createRoot((dispose) => {
           const initialMeta = meta[0].value
           const initialIcon = icon[0].value
-          const [mcpEnabled, setMcpEnabled] = createSignal(false)
 
           const [pathQuery, mcpQuery, lspQuery, providerQuery] = useQueries(() => ({
             queries: [
               input.queryOptions.path(key),
-              { ...input.queryOptions.mcp(key), enabled: mcpEnabled() },
+              input.queryOptions.mcp(key),
               input.queryOptions.lsp(key),
               input.queryOptions.providers(key),
             ],
@@ -242,7 +236,6 @@ export function createChildStoreManager(input: {
           })
           children[key] = child
           disposers.set(key, dispose)
-          mcpToggles.set(key, setMcpEnabled)
 
           const onPersistedInit = (init: Promise<string> | string | null, run: () => void) => {
             if (!(init instanceof Promise)) return
@@ -281,7 +274,6 @@ export function createChildStoreManager(input: {
     const key = directoryKey(directory)
     const childStore = ensureChild(directory)
     pinForOwner(key)
-    if (options.mcp) enableMcp(directory, key, childStore)
     const shouldBootstrap = options.bootstrap ?? true
     if (shouldBootstrap && childStore[0].status === "loading") {
       input.onBootstrap(directory)
@@ -292,7 +284,6 @@ export function createChildStoreManager(input: {
   function peek(directory: string, options: ChildOptions = {}) {
     const key = directoryKey(directory)
     const childStore = ensureChild(directory)
-    if (options.mcp) enableMcp(directory, key, childStore)
     const shouldBootstrap = options.bootstrap ?? true
     if (shouldBootstrap && childStore[0].status === "loading") {
       input.onBootstrap(directory)
@@ -300,19 +291,6 @@ export function createChildStoreManager(input: {
     return childStore
   }
 
-  function enableMcp(directory: string, key: DirectoryKey, childStore: [Store<State>, SetStoreFunction<State>]) {
-    if (mcpDirectories.has(key)) return
-    mcpDirectories.add(key)
-    mcpToggles.get(key)?.(true)
-    if (childStore[0].status !== "loading") input.onMcp(directory, childStore[1])
-  }
-
-  function disableMcp(directory: string) {
-    const key = directoryKey(directory)
-    if (!mcpDirectories.delete(key)) return
-    mcpToggles.get(key)?.(false)
-  }
-
   function projectMeta(directory: string, patch: ProjectMeta) {
     const key = directoryKey(directory)
     const [store, setStore] = ensureChild(directory)
@@ -352,8 +330,6 @@ export function createChildStoreManager(input: {
     pin,
     unpin,
     pinned,
-    mcp: (directory: string) => mcpDirectories.has(directoryKey(directory)),
-    disableMcp,
     disposeDirectory,
     runEviction,
     vcsCache,

packages/app/src/context/global-sync/types.ts

@@ -98,7 +98,6 @@ export type IconCache = {
 
 export type ChildOptions = {
   bootstrap?: boolean
-  mcp?: boolean
 }
 
 export type DirState = {

packages/app/src/context/server-sync.tsx

@@ -31,7 +31,6 @@ import { PathKey } from "@/utils/path-key"
 import { createDirSyncContext } from "./directory-sync"
 import { createSimpleContext, NormalizedProviderListResponse } from "@opencode-ai/ui/context"
 import { createRefCountMap } from "@/utils/refcount"
-import { retry } from "@opencode-ai/core/util/retry"
 
 type GlobalStore = {
   ready: boolean
@@ -206,19 +205,6 @@ export function createServerSyncContext() {
     onBootstrap: (directory) => {
       void bootstrapInstance(directory)
     },
-    onMcp: (directory, setStore) => {
-      void retry(() =>
-        sdkFor(directory)
-          .command.list()
-          .then((x) => setStore("command", x.data ?? [])),
-      ).catch((err) => {
-        showToast({
-          variant: "error",
-          title: language.t("toast.project.reloadFailed.title", { project: getFilename(directory) }),
-          description: formatServerError(err, language.t),
-        })
-      })
-    },
     onDispose: (directory) => {
       const key = directoryKey(directory)
       queue.clear(key)
@@ -325,7 +311,6 @@ export function createServerSyncContext() {
       const sdk = sdkFor(directory)
       await bootstrapDirectory({
         directory,
-        mcp: children.mcp(key),
         global: {
           config: globalStore.config,
           path: globalStore.path,
@@ -452,7 +437,6 @@ export function createServerSyncContext() {
     },
     child: children.child,
     peek: children.peek,
-    disableMcp: children.disableMcp,
     queryOptions: queryOptionsApi,
     // bootstrap,
     updateConfig: updateConfigMutation.mutateAsync,
@@ -470,11 +454,7 @@ export const { use: useServerSync, provider: ServerSyncProvider } = createSimple
 
     return {
       ...sync,
-      createDirSyncContext: createRefCountMap(
-        (dir) => createDirSyncContext(dir, sync),
-        (dir) => sync.disableMcp(dir),
-        directoryKey,
-      ),
+      createDirSyncContext: createRefCountMap((dir) => createDirSyncContext(dir, sync)),
     }
   },
 })

packages/app/src/utils/refcount.test.ts

@@ -1,49 +0,0 @@
-import { describe, expect, test } from "bun:test"
-import { createRoot } from "solid-js"
-import { createRefCountMap } from "./refcount"
-import { pathKey } from "./path-key"
-
-describe("createRefCountMap", () => {
-  test("removes an item after its last owner is disposed", () => {
-    const removed: string[] = []
-    const map = createRefCountMap(
-      (key) => key,
-      (key) => removed.push(key),
-    )
-    const first = createRoot((dispose) => {
-      map("/project")
-      return dispose
-    })
-    const second = createRoot((dispose) => {
-      map("/project")
-      return dispose
-    })
-
-    first()
-    expect(removed).toEqual([])
-    second()
-    expect(removed).toEqual(["/project"])
-  })
-
-  test("keeps equivalent path consumers until the last owner is disposed", () => {
-    const removed: string[] = []
-    const map = createRefCountMap(
-      (key) => key,
-      (key) => removed.push(key),
-      pathKey,
-    )
-    const first = createRoot((dispose) => {
-      map("C:\\repo")
-      return dispose
-    })
-    const second = createRoot((dispose) => {
-      map("C:/repo/")
-      return dispose
-    })
-
-    first()
-    expect(removed).toEqual([])
-    second()
-    expect(removed).toEqual(["C:/repo"])
-  })
-})

packages/app/src/utils/refcount.ts

@@ -1,32 +1,26 @@
 import { onCleanup } from "solid-js"
 
-export function createRefCountMap<T>(
-  create: (key: string) => T,
-  remove?: (key: string) => void,
-  identity: (key: string) => string = (key) => key,
-) {
+export function createRefCountMap<T>(create: (key: string) => T) {
   const items = new Map<string, T>()
   const refCounts = new Map<string, number>()
 
   return (key: string) => {
-    const id = identity(key)
     onCleanup(() => {
-      refCounts.set(id, (refCounts.get(id) ?? 0) - 1)
-      if (refCounts.get(id) === 0) {
-        remove?.(id)
-        items.delete(id)
-        refCounts.delete(id)
+      refCounts.set(key, (refCounts.get(key) ?? 0) - 1)
+      if (refCounts.get(key) === 0) {
+        items.delete(key)
+        refCounts.delete(key)
       }
     })
 
-    const cached = items.get(id)
+    const cached = items.get(key)
     if (cached) {
-      refCounts.set(id, (refCounts.get(id) ?? 0) + 1)
+      refCounts.set(key, (refCounts.get(key) ?? 0) + 1)
       return cached
     }
     const item = create(key)
-    items.set(id, item)
-    refCounts.set(id, 1)
+    items.set(key, item)
+    refCounts.set(key, 1)
     return item
   }
 }

packages/cli/package.json

@@ -1,25 +0,0 @@
-{
-  "$schema": "https://json.schemastore.org/package.json",
-  "name": "@opencode-ai/cli",
-  "version": "1.15.13",
-  "private": true,
-  "type": "module",
-  "license": "MIT",
-  "bin": {
-    "opencode": "./src/index.ts"
-  },
-  "scripts": {
-    "dev": "bun run src/index.ts",
-    "typecheck": "tsgo --noEmit"
-  },
-  "dependencies": {
-    "@effect/platform-node": "catalog:",
-    "@opencode-ai/core": "workspace:*",
-    "effect": "catalog:"
-  },
-  "devDependencies": {
-    "@tsconfig/bun": "catalog:",
-    "@types/bun": "catalog:",
-    "@typescript/native-preview": "catalog:"
-  }
-}

packages/cli/src/debug/agents.ts

@@ -1,19 +0,0 @@
-import { EOL } from "os"
-import { AgentV2 } from "@opencode-ai/core/agent"
-import { PluginBoot } from "@opencode-ai/core/plugin/boot"
-import * as Effect from "effect/Effect"
-import * as Command from "effect/unstable/cli/Command"
-
-export const AgentsCommand = Command.make("agents", {}, () =>
-  Effect.gen(function* () {
-    yield* PluginBoot.Service.use((service) => service.wait())
-    const agents = yield* AgentV2.Service.use((service) => service.all())
-    process.stdout.write(
-      JSON.stringify(
-        agents.sort((a, b) => a.id.localeCompare(b.id)),
-        null,
-        2,
-      ) + EOL,
-    )
-  }),
-).pipe(Command.withDescription("List all agents"))

packages/cli/src/debug/index.ts

@@ -1,7 +0,0 @@
-import * as Command from "effect/unstable/cli/Command"
-import { AgentsCommand } from "./agents"
-
-export const DebugCommand = Command.make("debug").pipe(
-  Command.withDescription("Debugging and troubleshooting tools"),
-  Command.withSubcommands([AgentsCommand]),
-)

packages/cli/src/index.ts

@@ -1,49 +0,0 @@
-#!/usr/bin/env bun
-
-import * as NodeRuntime from "@effect/platform-node/NodeRuntime"
-import * as NodeServices from "@effect/platform-node/NodeServices"
-import { AccountV2 } from "@opencode-ai/core/account"
-import { AgentV2 } from "@opencode-ai/core/agent"
-import { Catalog } from "@opencode-ai/core/catalog"
-import { Config } from "@opencode-ai/core/config"
-import { EventV2 } from "@opencode-ai/core/event"
-import { Location } from "@opencode-ai/core/location"
-import { Npm } from "@opencode-ai/core/npm"
-import { PluginV2 } from "@opencode-ai/core/plugin"
-import { PluginBoot } from "@opencode-ai/core/plugin/boot"
-import { Policy } from "@opencode-ai/core/policy"
-import { AbsolutePath } from "@opencode-ai/core/schema"
-import * as Effect from "effect/Effect"
-import * as Layer from "effect/Layer"
-import * as Command from "effect/unstable/cli/Command"
-import { DebugCommand } from "./debug"
-
-const cli = Command.make("opencode", {}, () => Effect.void).pipe(
-  Command.withDescription("OpenCode command line interface"),
-  Command.withSubcommands([DebugCommand]),
-)
-
-const locationLayer = Location.defaultLayer({
-  directory: AbsolutePath.make(process.cwd()),
-})
-
-const policyLayer = Policy.defaultLayer.pipe(Layer.provideMerge(locationLayer))
-const pluginLayer = PluginV2.defaultLayer
-const eventLayer = EventV2.defaultLayer
-
-const layer = PluginBoot.layer.pipe(
-  Layer.provideMerge(
-    Layer.mergeAll(
-      NodeServices.layer,
-      Catalog.layer.pipe(Layer.provideMerge(Layer.mergeAll(eventLayer, pluginLayer, policyLayer))),
-      eventLayer,
-      pluginLayer,
-      AccountV2.defaultLayer,
-      AgentV2.defaultLayer,
-      Config.defaultLayer.pipe(Layer.provideMerge(policyLayer)),
-      Npm.defaultLayer,
-    ),
-  ),
-)
-
-Command.run(cli, { version: "local" }).pipe(Effect.provide(layer), Effect.scoped, NodeRuntime.runMain)

packages/cli/tsconfig.json

@@ -1,7 +0,0 @@
-{
-  "$schema": "https://json.schemastore.org/tsconfig",
-  "extends": "@tsconfig/bun/tsconfig.json",
-  "compilerOptions": {
-    "noUncheckedIndexedAccess": false
-  }
-}

packages/console/app/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@opencode-ai/console-app",
-  "version": "1.15.13",
+  "version": "1.15.11",
   "type": "module",
   "license": "MIT",
   "scripts": {

packages/console/app/src/component/go-referral.tsx

@@ -2,6 +2,7 @@ import { action, json, query, useAction, useSubmission } from "@solidjs/router"
 import { createEffect, createMemo, createSignal, For, onCleanup, Show } from "solid-js"
 import { getRequestEvent } from "solid-js/web"
 import { Referral } from "@opencode-ai/console-core/referral.js"
+import { Actor } from "@opencode-ai/console-core/actor.js"
 import { withActor } from "~/context/auth.withActor"
 import { Modal } from "~/component/modal"
 import { IconCheck, IconCopy } from "~/component/icon"
@@ -9,7 +10,7 @@ import { useI18n } from "~/context/i18n"
 import { useLanguage } from "~/context/language"
 import { formatResetTime, liteResetTimeKeys } from "~/lib/format-reset-time"
 import { queryLiteSubscription } from "~/routes/workspace/[id]/go/lite-section"
-import { createReferralFromCookie } from "~/lib/referral-invite"
+import { clearReferralCookie, referralCodeFromCookieHeader } from "~/lib/referral-invite"
 import "./go-referral.css"
 
 type GoReferralSummary = Awaited<ReturnType<typeof Referral.summary>>
@@ -27,7 +28,18 @@ const emptyUsagePreview = {
 export const queryGoReferral = query(async (workspaceID: string) => {
   "use server"
   return withActor(async () => {
-    await createReferralFromCookie()
+    const event = getRequestEvent()
+    const referralCode = referralCodeFromCookieHeader(event?.request.headers.get("cookie") ?? null)
+    if (referralCode) {
+      await Referral.createFromAccount({
+        accountID: Actor.account(),
+        referralCode,
+      }).catch((error) => {
+        console.error("Referral create failed", error)
+      })
+      event?.response.headers.append("set-cookie", clearReferralCookie())
+    }
+
     return Referral.summary()
   }, workspaceID)
 }, "go.referral.get")

packages/console/app/src/i18n/ar.ts

@@ -249,7 +249,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | نماذج برمجة منخفضة التكلفة للجميع",
   "go.meta.description":
-    "يبدأ Go من $5 للشهر الأول، ثم $10/شهر، مع حدود طلب سخية لمدة 5 ساعات لـ GLM-5.1 وGLM-5 وKimi K2.5 وKimi K2.6 وMiMo-V2.5-Pro وMiMo-V2.5 وQwen3.7 Max وQwen3.6 Plus وMiniMax M2.5 وMiniMax M2.7 وDeepSeek V4 Pro وDeepSeek V4 Flash.",
+    "يبدأ Go من $5 للشهر الأول، ثم $10/شهر، مع حدود طلب سخية لمدة 5 ساعات لـ GLM-5.1 وGLM-5 وKimi K2.5 وKimi K2.6 وMiMo-V2.5-Pro وMiMo-V2.5 وQwen3.7 Max وQwen3.6 Plus وQwen3.5 Plus وMiniMax M2.5 وMiniMax M2.7 وDeepSeek V4 Pro وDeepSeek V4 Flash.",
   "go.hero.title": "نماذج برمجة منخفضة التكلفة للجميع",
   "go.hero.body":
     "يجلب Go البرمجة الوكيلة للمبرمجين حول العالم. يوفر حدودًا سخية ووصولًا موثوقًا إلى أقوى النماذج مفتوحة المصدر، حتى تتمكن من البناء باستخدام وكلاء أقوياء دون القلق بشأن التكلفة أو التوفر.",
@@ -298,7 +298,7 @@ export const dict = {
   "go.problem.item2": "حدود سخية ووصول موثوق",
   "go.problem.item3": "مصمم لأكبر عدد ممكن من المبرمجين",
   "go.problem.item4":
-    "يتضمن GLM-5.1 وGLM-5 وKimi K2.5 وKimi K2.6 وMiMo-V2.5-Pro وMiMo-V2.5 وQwen3.7 Max وQwen3.6 Plus وMiniMax M2.5 وMiniMax M2.7 وDeepSeek V4 Pro وDeepSeek V4 Flash",
+    "يتضمن GLM-5.1 وGLM-5 وKimi K2.5 وKimi K2.6 وMiMo-V2.5-Pro وMiMo-V2.5 وQwen3.7 Max وQwen3.6 Plus وQwen3.5 Plus وMiniMax M2.5 وMiniMax M2.7 وDeepSeek V4 Pro وDeepSeek V4 Flash",
   "go.how.title": "كيف يعمل Go",
   "go.how.body": "يبدأ Go من $5 للشهر الأول، ثم $10/شهر. يمكنك استخدامه مع OpenCode أو أي وكيل.",
   "go.how.step1.title": "أنشئ حسابًا",
@@ -322,7 +322,7 @@ export const dict = {
   "go.faq.a2": "يتضمن Go النماذج المدرجة أدناه، مع حدود سخية وإتاحة موثوقة.",
   "go.faq.q3": "هل Go هو نفسه Zen؟",
   "go.faq.a3":
-    "لا. Zen هو الدفع حسب الاستخدام، بينما يبدأ Go من $5 للشهر الأول، ثم $10/شهر، مع حدود سخية ووصول موثوق إلى نماذج المصدر المفتوح GLM-5.1 وGLM-5 وKimi K2.5 وKimi K2.6 وMiMo-V2.5-Pro وMiMo-V2.5 وQwen3.7 Max وQwen3.6 Plus وMiniMax M2.5 وMiniMax M2.7 وDeepSeek V4 Pro وDeepSeek V4 Flash.",
+    "لا. Zen هو الدفع حسب الاستخدام، بينما يبدأ Go من $5 للشهر الأول، ثم $10/شهر، مع حدود سخية ووصول موثوق إلى نماذج المصدر المفتوح GLM-5.1 وGLM-5 وKimi K2.5 وKimi K2.6 وMiMo-V2.5-Pro وMiMo-V2.5 وQwen3.7 Max وQwen3.6 Plus وQwen3.5 Plus وMiniMax M2.5 وMiniMax M2.7 وDeepSeek V4 Pro وDeepSeek V4 Flash.",
   "go.faq.q4": "كم تكلفة Go؟",
   "go.faq.a4.p1.beforePricing": "تكلفة Go",
   "go.faq.a4.p1.pricingLink": "$5 للشهر الأول",
@@ -345,7 +345,7 @@ export const dict = {
 
   "go.faq.q9": "ما الفرق بين النماذج المجانية وGo؟",
   "go.faq.a9":
-    "تشمل النماذج المجانية Big Pickle بالإضافة إلى النماذج الترويجية المتاحة في ذلك الوقت، مع حصة 200 طلب/يوم. يتضمن Go نماذج GLM-5.1 وGLM-5 وKimi K2.5 وKimi K2.6 وMiMo-V2.5-Pro وMiMo-V2.5 وQwen3.7 Max وQwen3.6 Plus وMiniMax M2.5 وMiniMax M2.7 وDeepSeek V4 Pro وDeepSeek V4 Flash مع حصص طلبات أعلى مطبقة عبر نوافذ متجددة (5 ساعات، أسبوعيًا، وشهريًا)، تعادل تقريبًا 12 دولارًا كل 5 ساعات، و30 دولارًا في الأسبوع، و60 دولارًا في الشهر (تختلف أعداد الطلبات الفعلية حسب النموذج والاستخدام).",
+    "تشمل النماذج المجانية Big Pickle بالإضافة إلى النماذج الترويجية المتاحة في ذلك الوقت، مع حصة 200 طلب/يوم. يتضمن Go نماذج GLM-5.1 وGLM-5 وKimi K2.5 وKimi K2.6 وMiMo-V2.5-Pro وMiMo-V2.5 وQwen3.7 Max وQwen3.6 Plus وQwen3.5 Plus وMiniMax M2.5 وMiniMax M2.7 وDeepSeek V4 Pro وDeepSeek V4 Flash مع حصص طلبات أعلى مطبقة عبر نوافذ متجددة (5 ساعات، أسبوعيًا، وشهريًا)، تعادل تقريبًا 12 دولارًا كل 5 ساعات، و30 دولارًا في الأسبوع، و60 دولارًا في الشهر (تختلف أعداد الطلبات الفعلية حسب النموذج والاستخدام).",
 
   "zen.api.error.rateLimitExceeded": "تم تجاوز حد الطلبات. يرجى المحاولة مرة أخرى لاحقًا.",
   "zen.api.error.modelNotSupported": "النموذج {{model}} غير مدعوم",

packages/console/app/src/i18n/br.ts

@@ -253,7 +253,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Modelos de codificação de baixo custo para todos",
   "go.meta.description":
-    "O Go começa em $5 no primeiro mês, depois $10/mês, com limites generosos de solicitação de 5 horas para GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash.",
+    "O Go começa em $5 no primeiro mês, depois $10/mês, com limites generosos de solicitação de 5 horas para GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash.",
   "go.hero.title": "Modelos de codificação de baixo custo para todos",
   "go.hero.body":
     "O Go traz a codificação com agentes para programadores em todo o mundo. Oferecendo limites generosos e acesso confiável aos modelos de código aberto mais capazes, para que você possa construir com agentes poderosos sem se preocupar com custos ou disponibilidade.",
@@ -303,7 +303,7 @@ export const dict = {
   "go.problem.item2": "Limites generosos e acesso confiável",
   "go.problem.item3": "Feito para o maior número possível de programadores",
   "go.problem.item4":
-    "Inclui GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash",
+    "Inclui GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash",
   "go.how.title": "Como o Go funciona",
   "go.how.body":
     "O Go começa em $5 no primeiro mês, depois $10/mês. Você pode usá-lo com o OpenCode ou qualquer agente.",
@@ -329,7 +329,7 @@ export const dict = {
   "go.faq.a2": "O Go inclui os modelos listados abaixo, com limites generosos e acesso confiável.",
   "go.faq.q3": "O Go é o mesmo que o Zen?",
   "go.faq.a3":
-    "Não. Zen é pay-as-you-go, enquanto o Go começa em $5 no primeiro mês, depois $10/mês, com limites generosos e acesso confiável aos modelos open source GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash.",
+    "Não. Zen é pay-as-you-go, enquanto o Go começa em $5 no primeiro mês, depois $10/mês, com limites generosos e acesso confiável aos modelos open source GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash.",
   "go.faq.q4": "Quanto custa o Go?",
   "go.faq.a4.p1.beforePricing": "O Go custa",
   "go.faq.a4.p1.pricingLink": "$5 no primeiro mês",
@@ -353,7 +353,7 @@ export const dict = {
 
   "go.faq.q9": "Qual a diferença entre os modelos gratuitos e o Go?",
   "go.faq.a9":
-    "Os modelos gratuitos incluem Big Pickle e modelos promocionais disponíveis no momento, com uma cota de 200 requisições/dia. O Go inclui GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash com cotas de requisição mais altas aplicadas em janelas móveis (5 horas, semanal e mensal), aproximadamente equivalentes a $12 por 5 horas, $30 por semana e $60 por mês (as contagens reais de requisições variam de acordo com o modelo e o uso).",
+    "Os modelos gratuitos incluem Big Pickle e modelos promocionais disponíveis no momento, com uma cota de 200 requisições/dia. O Go inclui GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash com cotas de requisição mais altas aplicadas em janelas móveis (5 horas, semanal e mensal), aproximadamente equivalentes a $12 por 5 horas, $30 por semana e $60 por mês (as contagens reais de requisições variam de acordo com o modelo e o uso).",
 
   "zen.api.error.rateLimitExceeded": "Limite de taxa excedido. Por favor, tente novamente mais tarde.",
   "zen.api.error.modelNotSupported": "Modelo {{model}} não suportado",

packages/console/app/src/i18n/da.ts

@@ -251,7 +251,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Kodningsmodeller til lav pris for alle",
   "go.meta.description":
-    "Go starter ved $5 for den første måned, derefter $10/måned, med generøse 5-timers anmodningsgrænser for GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash.",
+    "Go starter ved $5 for den første måned, derefter $10/måned, med generøse 5-timers anmodningsgrænser for GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash.",
   "go.hero.title": "Kodningsmodeller til lav pris for alle",
   "go.hero.body":
     "Go bringer agentisk kodning til programmører over hele verden. Med generøse grænser og pålidelig adgang til de mest kapable open source-modeller, så du kan bygge med kraftfulde agenter uden at bekymre dig om omkostninger eller tilgængelighed.",
@@ -300,7 +300,7 @@ export const dict = {
   "go.problem.item2": "Generøse grænser og pålidelig adgang",
   "go.problem.item3": "Bygget til så mange programmører som muligt",
   "go.problem.item4":
-    "Inkluderer GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash",
+    "Inkluderer GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash",
   "go.how.title": "Hvordan Go virker",
   "go.how.body":
     "Go starter ved $5 for den første måned, derefter $10/måned. Du kan bruge det med OpenCode eller enhver agent.",
@@ -326,7 +326,7 @@ export const dict = {
   "go.faq.a2": "Go inkluderer modellerne nedenfor med generøse grænser og pålidelig adgang.",
   "go.faq.q3": "Er Go det samme som Zen?",
   "go.faq.a3":
-    "Nej. Zen er pay-as-you-go, mens Go starter ved $5 for den første måned, derefter $10/måned, med generøse grænser og pålidelig adgang til open source-modellerne GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash.",
+    "Nej. Zen er pay-as-you-go, mens Go starter ved $5 for den første måned, derefter $10/måned, med generøse grænser og pålidelig adgang til open source-modellerne GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash.",
   "go.faq.q4": "Hvad koster Go?",
   "go.faq.a4.p1.beforePricing": "Go koster",
   "go.faq.a4.p1.pricingLink": "$5 første måned",
@@ -349,7 +349,7 @@ export const dict = {
 
   "go.faq.q9": "Hvad er forskellen på gratis modeller og Go?",
   "go.faq.a9":
-    "Gratis modeller inkluderer Big Pickle plus salgsfremmende modeller tilgængelige på det tidspunkt, med en kvote på 200 forespørgsler/dag. Go inkluderer GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash med højere anmodningskvoter håndhævet over rullende vinduer (5-timers, ugentlig og månedlig), nogenlunde svarende til $12 pr. 5 timer, $30 pr. uge og $60 pr. måned (faktiske anmodningstal varierer efter model og brug).",
+    "Gratis modeller inkluderer Big Pickle plus salgsfremmende modeller tilgængelige på det tidspunkt, med en kvote på 200 forespørgsler/dag. Go inkluderer GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash med højere anmodningskvoter håndhævet over rullende vinduer (5-timers, ugentlig og månedlig), nogenlunde svarende til $12 pr. 5 timer, $30 pr. uge og $60 pr. måned (faktiske anmodningstal varierer efter model og brug).",
 
   "zen.api.error.rateLimitExceeded": "Hastighedsgrænse overskredet. Prøv venligst igen senere.",
   "zen.api.error.modelNotSupported": "Model {{model}} understøttes ikke",

packages/console/app/src/i18n/de.ts

@@ -253,7 +253,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Kostengünstige Coding-Modelle für alle",
   "go.meta.description":
-    "Go beginnt bei $5 für den ersten Monat, danach $10/Monat, mit großzügigen 5-Stunden-Anfragelimits für GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro und DeepSeek V4 Flash.",
+    "Go beginnt bei $5 für den ersten Monat, danach $10/Monat, mit großzügigen 5-Stunden-Anfragelimits für GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro und DeepSeek V4 Flash.",
   "go.hero.title": "Kostengünstige Coding-Modelle für alle",
   "go.hero.body":
     "Go bringt Agentic Coding zu Programmierern auf der ganzen Welt. Mit großzügigen Limits und zuverlässigem Zugang zu den leistungsfähigsten Open-Source-Modellen, damit du mit leistungsstarken Agenten entwickeln kannst, ohne dir Gedanken über Kosten oder Verfügbarkeit zu machen.",
@@ -302,7 +302,7 @@ export const dict = {
   "go.problem.item2": "Großzügige Limits und zuverlässiger Zugang",
   "go.problem.item3": "Für so viele Programmierer wie möglich gebaut",
   "go.problem.item4":
-    "Beinhaltet GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro und DeepSeek V4 Flash",
+    "Beinhaltet GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro und DeepSeek V4 Flash",
   "go.how.title": "Wie Go funktioniert",
   "go.how.body":
     "Go beginnt bei $5 für den ersten Monat, danach $10/Monat. Du kannst es mit OpenCode oder jedem Agenten nutzen.",
@@ -328,7 +328,7 @@ export const dict = {
   "go.faq.a2": "Go umfasst die unten aufgeführten Modelle mit großzügigen Limits und zuverlässigem Zugriff.",
   "go.faq.q3": "Ist Go dasselbe wie Zen?",
   "go.faq.a3":
-    "Nein. Zen ist Pay-as-you-go, während Go bei $5 für den ersten Monat beginnt, danach $10/Monat, mit großzügigen Limits und zuverlässigem Zugang zu den Open-Source-Modellen GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro und DeepSeek V4 Flash.",
+    "Nein. Zen ist Pay-as-you-go, während Go bei $5 für den ersten Monat beginnt, danach $10/Monat, mit großzügigen Limits und zuverlässigem Zugang zu den Open-Source-Modellen GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro und DeepSeek V4 Flash.",
   "go.faq.q4": "Wie viel kostet Go?",
   "go.faq.a4.p1.beforePricing": "Go kostet",
   "go.faq.a4.p1.pricingLink": "$5 im ersten Monat",
@@ -352,7 +352,7 @@ export const dict = {
 
   "go.faq.q9": "Was ist der Unterschied zwischen kostenlosen Modellen und Go?",
   "go.faq.a9":
-    "Kostenlose Modelle beinhalten Big Pickle sowie Werbemodelle, die zum jeweiligen Zeitpunkt verfügbar sind, mit einem Kontingent von 200 Anfragen/Tag. Go beinhaltet GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro und DeepSeek V4 Flash mit höheren Anfragekontingenten, die über rollierende Zeitfenster (5 Stunden, wöchentlich und monatlich) durchgesetzt werden, grob äquivalent zu $12 pro 5 Stunden, $30 pro Woche und $60 pro Monat (tatsächliche Anfragezahlen variieren je nach Modell und Nutzung).",
+    "Kostenlose Modelle beinhalten Big Pickle sowie Werbemodelle, die zum jeweiligen Zeitpunkt verfügbar sind, mit einem Kontingent von 200 Anfragen/Tag. Go beinhaltet GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro und DeepSeek V4 Flash mit höheren Anfragekontingenten, die über rollierende Zeitfenster (5 Stunden, wöchentlich und monatlich) durchgesetzt werden, grob äquivalent zu $12 pro 5 Stunden, $30 pro Woche und $60 pro Monat (tatsächliche Anfragezahlen variieren je nach Modell und Nutzung).",
 
   "zen.api.error.rateLimitExceeded": "Ratenlimit überschritten. Bitte versuche es später erneut.",
   "zen.api.error.modelNotSupported": "Modell {{model}} wird nicht unterstützt",

packages/console/app/src/i18n/en.ts

@@ -248,7 +248,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Low cost coding models for everyone",
   "go.meta.description":
-    "Go starts at $5 for your first month, then $10/month, with generous 5-hour request limits for GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, and DeepSeek V4 Flash.",
+    "Go starts at $5 for your first month, then $10/month, with generous 5-hour request limits for GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, and DeepSeek V4 Flash.",
   "go.hero.title": "Low cost coding models for everyone",
   "go.hero.body":
     "Go brings agentic coding to programmers around the world. Offering generous limits and reliable access to the most capable open-source models, so you can build with powerful agents without worrying about cost or availability.",
@@ -296,7 +296,7 @@ export const dict = {
   "go.problem.item2": "Generous limits and reliable access",
   "go.problem.item3": "Built for as many programmers as possible",
   "go.problem.item4":
-    "Includes GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, and DeepSeek V4 Flash",
+    "Includes GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, and DeepSeek V4 Flash",
   "go.how.title": "How Go works",
   "go.how.body": "Go starts at $5 for your first month, then $10/month. You can use it with OpenCode or any agent.",
   "go.how.step1.title": "Create an account",
@@ -321,7 +321,7 @@ export const dict = {
   "go.faq.a2": "Go includes the models listed below, with generous limits and reliable access.",
   "go.faq.q3": "Is Go the same as Zen?",
   "go.faq.a3":
-    "No. Zen is pay-as-you-go, while Go starts at $5 for your first month, then $10/month, with generous limits and reliable access to open-source models GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, and DeepSeek V4 Flash.",
+    "No. Zen is pay-as-you-go, while Go starts at $5 for your first month, then $10/month, with generous limits and reliable access to open-source models GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, and DeepSeek V4 Flash.",
   "go.faq.q4": "How much does Go cost?",
   "go.faq.a4.p1.beforePricing": "Go costs",
   "go.faq.a4.p1.pricingLink": "$5 first month",
@@ -345,7 +345,7 @@ export const dict = {
 
   "go.faq.q9": "What is the difference between free models and Go?",
   "go.faq.a9":
-    "Free models include Big Pickle plus promotional models available at the time, with a quota of 200 requests/day. Go includes GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, and DeepSeek V4 Flash with higher request quotas enforced across rolling windows (5-hour, weekly, and monthly), roughly equivalent to $12 per 5 hours, $30 per week, and $60 per month (actual request counts vary by model and usage).",
+    "Free models include Big Pickle plus promotional models available at the time, with a quota of 200 requests/day. Go includes GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, and DeepSeek V4 Flash with higher request quotas enforced across rolling windows (5-hour, weekly, and monthly), roughly equivalent to $12 per 5 hours, $30 per week, and $60 per month (actual request counts vary by model and usage).",
 
   "zen.api.error.rateLimitExceeded": "Rate limit exceeded. Please try again later.",
   "zen.api.error.modelNotSupported": "Model {{model}} is not supported",

packages/console/app/src/i18n/es.ts

@@ -254,7 +254,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Modelos de programación de bajo coste para todos",
   "go.meta.description":
-    "Go comienza en $5 el primer mes, luego 10 $/mes, con generosos límites de solicitudes de 5 horas para GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro y DeepSeek V4 Flash.",
+    "Go comienza en $5 el primer mes, luego 10 $/mes, con generosos límites de solicitudes de 5 horas para GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro y DeepSeek V4 Flash.",
   "go.hero.title": "Modelos de programación de bajo coste para todos",
   "go.hero.body":
     "Go lleva la programación agéntica a programadores de todo el mundo. Ofrece límites generosos y acceso fiable a los modelos de código abierto más capaces, para que puedas crear con agentes potentes sin preocuparte por el coste o la disponibilidad.",
@@ -304,7 +304,7 @@ export const dict = {
   "go.problem.item2": "Límites generosos y acceso fiable",
   "go.problem.item3": "Creado para tantos programadores como sea posible",
   "go.problem.item4":
-    "Incluye GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro y DeepSeek V4 Flash",
+    "Incluye GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro y DeepSeek V4 Flash",
   "go.how.title": "Cómo funciona Go",
   "go.how.body": "Go comienza en $5 el primer mes, luego 10 $/mes. Puedes usarlo con OpenCode o cualquier agente.",
   "go.how.step1.title": "Crear una cuenta",
@@ -329,7 +329,7 @@ export const dict = {
   "go.faq.a2": "Go incluye los modelos que se indican abajo, con límites generosos y acceso confiable.",
   "go.faq.q3": "¿Es Go lo mismo que Zen?",
   "go.faq.a3":
-    "No. Zen es pago por uso, mientras que Go comienza en $5 el primer mes, luego 10 $/mes, con límites generosos y acceso fiable a los modelos de código abierto GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro y DeepSeek V4 Flash.",
+    "No. Zen es pago por uso, mientras que Go comienza en $5 el primer mes, luego 10 $/mes, con límites generosos y acceso fiable a los modelos de código abierto GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro y DeepSeek V4 Flash.",
   "go.faq.q4": "¿Cuánto cuesta Go?",
   "go.faq.a4.p1.beforePricing": "Go cuesta",
   "go.faq.a4.p1.pricingLink": "$5 el primer mes",
@@ -353,7 +353,7 @@ export const dict = {
 
   "go.faq.q9": "¿Cuál es la diferencia entre los modelos gratuitos y Go?",
   "go.faq.a9":
-    "Los modelos gratuitos incluyen Big Pickle más modelos promocionales disponibles en el momento, con una cuota de 200 solicitudes/día. Go incluye GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro y DeepSeek V4 Flash con cuotas de solicitud más altas aplicadas a través de ventanas móviles (5 horas, semanal y mensual), aproximadamente equivalente a 12 $ por 5 horas, 30 $ por semana y 60 $ por mes (los recuentos reales de solicitudes varían según el modelo y el uso).",
+    "Los modelos gratuitos incluyen Big Pickle más modelos promocionales disponibles en el momento, con una cuota de 200 solicitudes/día. Go incluye GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro y DeepSeek V4 Flash con cuotas de solicitud más altas aplicadas a través de ventanas móviles (5 horas, semanal y mensual), aproximadamente equivalente a 12 $ por 5 horas, 30 $ por semana y 60 $ por mes (los recuentos reales de solicitudes varían según el modelo y el uso).",
 
   "zen.api.error.rateLimitExceeded": "Límite de tasa excedido. Por favor, inténtalo de nuevo más tarde.",
   "zen.api.error.modelNotSupported": "Modelo {{model}} no soportado",

packages/console/app/src/i18n/fr.ts

@@ -255,7 +255,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Modèles de code à faible coût pour tous",
   "go.meta.description":
-    "Go commence à $5 pour le premier mois, puis 10 $/mois, avec des limites de requêtes généreuses sur 5 heures pour GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro et DeepSeek V4 Flash.",
+    "Go commence à $5 pour le premier mois, puis 10 $/mois, avec des limites de requêtes généreuses sur 5 heures pour GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro et DeepSeek V4 Flash.",
   "go.hero.title": "Modèles de code à faible coût pour tous",
   "go.hero.body":
     "Go apporte le codage agentique aux programmeurs du monde entier. Offrant des limites généreuses et un accès fiable aux modèles open source les plus capables, pour que vous puissiez construire avec des agents puissants sans vous soucier du coût ou de la disponibilité.",
@@ -304,7 +304,7 @@ export const dict = {
   "go.problem.item2": "Limites généreuses et accès fiable",
   "go.problem.item3": "Conçu pour autant de programmeurs que possible",
   "go.problem.item4":
-    "Inclut GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro et DeepSeek V4 Flash",
+    "Inclut GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro et DeepSeek V4 Flash",
   "go.how.title": "Comment fonctionne Go",
   "go.how.body":
     "Go commence à $5 pour le premier mois, puis 10 $/mois. Vous pouvez l'utiliser avec OpenCode ou n'importe quel agent.",
@@ -330,7 +330,7 @@ export const dict = {
   "go.faq.a2": "Go inclut les modèles ci-dessous, avec des limites généreuses et un accès fiable.",
   "go.faq.q3": "Est-ce que Go est la même chose que Zen ?",
   "go.faq.a3":
-    "Non. Zen est un paiement à l'utilisation, tandis que Go commence à $5 pour le premier mois, puis 10 $/mois, avec des limites généreuses et un accès fiable aux modèles open source GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro et DeepSeek V4 Flash.",
+    "Non. Zen est un paiement à l'utilisation, tandis que Go commence à $5 pour le premier mois, puis 10 $/mois, avec des limites généreuses et un accès fiable aux modèles open source GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro et DeepSeek V4 Flash.",
   "go.faq.q4": "Combien coûte Go ?",
   "go.faq.a4.p1.beforePricing": "Go coûte",
   "go.faq.a4.p1.pricingLink": "$5 le premier mois",
@@ -353,7 +353,7 @@ export const dict = {
     "Oui, vous pouvez utiliser Go avec n'importe quel agent. Suivez les instructions de configuration dans votre agent de code préféré.",
   "go.faq.q9": "Quelle est la différence entre les modèles gratuits et Go ?",
   "go.faq.a9":
-    "Les modèles gratuits incluent Big Pickle ainsi que des modèles promotionnels disponibles à ce moment-là, avec un quota de 200 requêtes/jour. Go inclut GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro et DeepSeek V4 Flash avec des quotas de requêtes plus élevés appliqués sur des fenêtres glissantes (5 heures, hebdomadaire et mensuelle), à peu près équivalent à 12 $ par 5 heures, 30 $ par semaine et 60 $ par mois (le nombre réel de requêtes varie selon le modèle et l'utilisation).",
+    "Les modèles gratuits incluent Big Pickle ainsi que des modèles promotionnels disponibles à ce moment-là, avec un quota de 200 requêtes/jour. Go inclut GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro et DeepSeek V4 Flash avec des quotas de requêtes plus élevés appliqués sur des fenêtres glissantes (5 heures, hebdomadaire et mensuelle), à peu près équivalent à 12 $ par 5 heures, 30 $ par semaine et 60 $ par mois (le nombre réel de requêtes varie selon le modèle et l'utilisation).",
 
   "zen.api.error.rateLimitExceeded": "Limite de débit dépassée. Veuillez réessayer plus tard.",
   "zen.api.error.modelNotSupported": "Modèle {{model}} non pris en charge",

packages/console/app/src/i18n/it.ts

@@ -251,7 +251,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Modelli di coding a basso costo per tutti",
   "go.meta.description":
-    "Go inizia a $5 per il primo mese, poi $10/mese, con generosi limiti di richiesta di 5 ore per GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash.",
+    "Go inizia a $5 per il primo mese, poi $10/mese, con generosi limiti di richiesta di 5 ore per GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash.",
   "go.hero.title": "Modelli di coding a basso costo per tutti",
   "go.hero.body":
     "Go porta il coding agentico ai programmatori di tutto il mondo. Offrendo limiti generosi e un accesso affidabile ai modelli open source più capaci, in modo da poter costruire con agenti potenti senza preoccuparsi dei costi o della disponibilità.",
@@ -300,7 +300,7 @@ export const dict = {
   "go.problem.item2": "Limiti generosi e accesso affidabile",
   "go.problem.item3": "Costruito per il maggior numero possibile di programmatori",
   "go.problem.item4":
-    "Include GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash",
+    "Include GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash",
   "go.how.title": "Come funziona Go",
   "go.how.body": "Go inizia a $5 per il primo mese, poi $10/mese. Puoi usarlo con OpenCode o qualsiasi agente.",
   "go.how.step1.title": "Crea un account",
@@ -325,7 +325,7 @@ export const dict = {
   "go.faq.a2": "Go include i modelli elencati di seguito, con limiti generosi e accesso affidabile.",
   "go.faq.q3": "Go è lo stesso di Zen?",
   "go.faq.a3":
-    "No. Zen è a consumo, mentre Go inizia a $5 per il primo mese, poi $10/mese, con limiti generosi e accesso affidabile ai modelli open source GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash.",
+    "No. Zen è a consumo, mentre Go inizia a $5 per il primo mese, poi $10/mese, con limiti generosi e accesso affidabile ai modelli open source GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash.",
   "go.faq.q4": "Quanto costa Go?",
   "go.faq.a4.p1.beforePricing": "Go costa",
   "go.faq.a4.p1.pricingLink": "$5 il primo mese",
@@ -349,7 +349,7 @@ export const dict = {
 
   "go.faq.q9": "Qual è la differenza tra i modelli gratuiti e Go?",
   "go.faq.a9":
-    "I modelli gratuiti includono Big Pickle più modelli promozionali disponibili al momento, con una quota di 200 richieste/giorno. Go include GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash con quote di richiesta più elevate applicate su finestre mobili (5 ore, settimanale e mensile), approssimativamente equivalenti a $12 ogni 5 ore, $30 a settimana e $60 al mese (il conteggio effettivo delle richieste varia in base al modello e all'utilizzo).",
+    "I modelli gratuiti includono Big Pickle più modelli promozionali disponibili al momento, con una quota di 200 richieste/giorno. Go include GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro e DeepSeek V4 Flash con quote di richiesta più elevate applicate su finestre mobili (5 ore, settimanale e mensile), approssimativamente equivalenti a $12 ogni 5 ore, $30 a settimana e $60 al mese (il conteggio effettivo delle richieste varia in base al modello e all'utilizzo).",
 
   "zen.api.error.rateLimitExceeded": "Limite di richieste superato. Riprova più tardi.",
   "zen.api.error.modelNotSupported": "Modello {{model}} non supportato",

packages/console/app/src/i18n/ja.ts

@@ -250,7 +250,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | すべての人のための低価格なコーディングモデル",
   "go.meta.description":
-    "Goは最初の月$5、その後$10/月で、GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro、DeepSeek V4 Flashに対して5時間のゆとりあるリクエスト上限があります。",
+    "Goは最初の月$5、その後$10/月で、GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、Qwen3.5 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro、DeepSeek V4 Flashに対して5時間のゆとりあるリクエスト上限があります。",
   "go.hero.title": "すべての人のための低価格なコーディングモデル",
   "go.hero.body":
     "Goは、世界中のプログラマーにエージェント型コーディングをもたらします。最も高性能なオープンソースモデルへの十分な制限と安定したアクセスを提供し、コストや可用性を気にすることなく強力なエージェントで構築できます。",
@@ -300,7 +300,7 @@ export const dict = {
   "go.problem.item2": "十分な制限と安定したアクセス",
   "go.problem.item3": "できるだけ多くのプログラマーのために構築",
   "go.problem.item4":
-    "GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro、DeepSeek V4 Flashを含む",
+    "GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、Qwen3.5 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro、DeepSeek V4 Flashを含む",
   "go.how.title": "Goの仕組み",
   "go.how.body": "Goは最初の月$5、その後$10/月で始まります。OpenCodeまたは任意のエージェントで使えます。",
   "go.how.step1.title": "アカウントを作成",
@@ -325,7 +325,7 @@ export const dict = {
   "go.faq.a2": "Go には、十分な利用上限と安定したアクセスを備えた、以下のモデルが含まれます。",
   "go.faq.q3": "GoはZenと同じですか？",
   "go.faq.a3":
-    "いいえ。Zenは従量課金制ですが、Goは最初の月$5、その後$10/月で始まり、GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro、DeepSeek V4 Flashのオープンソースモデルに対して、ゆとりある上限と信頼できるアクセスを提供します。",
+    "いいえ。Zenは従量課金制ですが、Goは最初の月$5、その後$10/月で始まり、GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、Qwen3.5 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro、DeepSeek V4 Flashのオープンソースモデルに対して、ゆとりある上限と信頼できるアクセスを提供します。",
   "go.faq.q4": "Goの料金は？",
   "go.faq.a4.p1.beforePricing": "Goは",
   "go.faq.a4.p1.pricingLink": "最初の月$5",
@@ -349,7 +349,7 @@ export const dict = {
 
   "go.faq.q9": "無料モデルとGoの違いは何ですか？",
   "go.faq.a9":
-    "無料モデルにはBig Pickleと、その時点で利用可能なプロモーションモデルが含まれ、1日200リクエストの制限があります。GoにはGLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro、DeepSeek V4 Flashが含まれ、ローリングウィンドウ（5時間、週間、月間）全体でより高いリクエスト制限が適用されます。これは概算で5時間あたり$12、週間$30、月間$60相当です（実際のリクエスト数はモデルと使用状況により異なります）。",
+    "無料モデルにはBig Pickleと、その時点で利用可能なプロモーションモデルが含まれ、1日200リクエストの制限があります。GoにはGLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、Qwen3.5 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro、DeepSeek V4 Flashが含まれ、ローリングウィンドウ（5時間、週間、月間）全体でより高いリクエスト制限が適用されます。これは概算で5時間あたり$12、週間$30、月間$60相当です（実際のリクエスト数はモデルと使用状況により異なります）。",
 
   "zen.api.error.rateLimitExceeded": "レート制限を超えました。後でもう一度お試しください。",
   "zen.api.error.modelNotSupported": "モデル {{model}} はサポートされていません",

packages/console/app/src/i18n/ko.ts

@@ -247,7 +247,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | 모두를 위한 저비용 코딩 모델",
   "go.meta.description":
-    "Go는 첫 달 $5, 이후 $10/월로 시작하며, GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash에 대해 넉넉한 5시간 요청 한도를 제공합니다.",
+    "Go는 첫 달 $5, 이후 $10/월로 시작하며, GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash에 대해 넉넉한 5시간 요청 한도를 제공합니다.",
   "go.hero.title": "모두를 위한 저비용 코딩 모델",
   "go.hero.body":
     "Go는 전 세계 프로그래머들에게 에이전트 코딩을 제공합니다. 가장 유능한 오픈 소스 모델에 대한 넉넉한 한도와 안정적인 액세스를 제공하므로, 비용이나 가용성 걱정 없이 강력한 에이전트로 빌드할 수 있습니다.",
@@ -297,7 +297,7 @@ export const dict = {
   "go.problem.item2": "넉넉한 한도와 안정적인 액세스",
   "go.problem.item3": "가능한 한 많은 프로그래머를 위해 제작됨",
   "go.problem.item4":
-    "GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash 포함",
+    "GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash 포함",
   "go.how.title": "Go 작동 방식",
   "go.how.body": "Go는 첫 달 $5, 이후 $10/월로 시작합니다. OpenCode 또는 어떤 에이전트와도 함께 사용할 수 있습니다.",
   "go.how.step1.title": "계정 생성",
@@ -321,7 +321,7 @@ export const dict = {
   "go.faq.a2": "Go에는 넉넉한 한도와 안정적인 액세스를 제공하는 아래 모델이 포함됩니다.",
   "go.faq.q3": "Go는 Zen과 같은가요?",
   "go.faq.a3":
-    "아니요. Zen은 종량제인 반면, Go는 첫 달 $5, 이후 $10/월로 시작하며, GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash 오픈 소스 모델에 대한 넉넉한 한도와 안정적인 액세스를 제공합니다.",
+    "아니요. Zen은 종량제인 반면, Go는 첫 달 $5, 이후 $10/월로 시작하며, GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash 오픈 소스 모델에 대한 넉넉한 한도와 안정적인 액세스를 제공합니다.",
   "go.faq.q4": "Go 비용은 얼마인가요?",
   "go.faq.a4.p1.beforePricing": "Go 비용은",
   "go.faq.a4.p1.pricingLink": "첫 달 $5",
@@ -344,7 +344,7 @@ export const dict = {
 
   "go.faq.q9": "무료 모델과 Go의 차이점은 무엇인가요?",
   "go.faq.a9":
-    "무료 모델에는 Big Pickle과 당시 사용 가능한 프로모션 모델이 포함되며, 하루 200회 요청 할당량이 적용됩니다. Go는 GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash를 포함하며, 롤링 윈도우(5시간, 주간, 월간)에 걸쳐 더 높은 요청 할당량을 적용합니다. 이는 대략 5시간당 $12, 주당 $30, 월 $60에 해당합니다(실제 요청 수는 모델 및 사용량에 따라 다름).",
+    "무료 모델에는 Big Pickle과 당시 사용 가능한 프로모션 모델이 포함되며, 하루 200회 요청 할당량이 적용됩니다. Go는 GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash를 포함하며, 롤링 윈도우(5시간, 주간, 월간)에 걸쳐 더 높은 요청 할당량을 적용합니다. 이는 대략 5시간당 $12, 주당 $30, 월 $60에 해당합니다(실제 요청 수는 모델 및 사용량에 따라 다름).",
 
   "zen.api.error.rateLimitExceeded": "속도 제한을 초과했습니다. 나중에 다시 시도해 주세요.",
   "zen.api.error.modelNotSupported": "{{model}} 모델은 지원되지 않습니다",

packages/console/app/src/i18n/no.ts

@@ -251,7 +251,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Rimelige kodemodeller for alle",
   "go.meta.description":
-    "Go starter på $5 for den første måneden, deretter $10/måned, med sjenerøse 5-timers forespørselsgrenser for GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash.",
+    "Go starter på $5 for den første måneden, deretter $10/måned, med sjenerøse 5-timers forespørselsgrenser for GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash.",
   "go.hero.title": "Rimelige kodemodeller for alle",
   "go.hero.body":
     "Go bringer agent-koding til programmerere over hele verden. Med rause grenser og pålitelig tilgang til de mest kapable åpen kildekode-modellene, kan du bygge med kraftige agenter uten å bekymre deg for kostnader eller tilgjengelighet.",
@@ -300,7 +300,7 @@ export const dict = {
   "go.problem.item2": "Rause grenser og pålitelig tilgang",
   "go.problem.item3": "Bygget for så mange programmerere som mulig",
   "go.problem.item4":
-    "Inkluderer GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash",
+    "Inkluderer GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash",
   "go.how.title": "Hvordan Go fungerer",
   "go.how.body":
     "Go starter på $5 for den første måneden, deretter $10/måned. Du kan bruke det med OpenCode eller hvilken som helst agent.",
@@ -326,7 +326,7 @@ export const dict = {
   "go.faq.a2": "Go inkluderer modellene nedenfor, med høye grenser og pålitelig tilgang.",
   "go.faq.q3": "Er Go det samme som Zen?",
   "go.faq.a3":
-    "Nei. Zen er betaling etter bruk, mens Go starter på $5 for den første måneden, deretter $10/måned, med sjenerøse grenser og pålitelig tilgang til åpen kildekode-modellene GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash.",
+    "Nei. Zen er betaling etter bruk, mens Go starter på $5 for den første måneden, deretter $10/måned, med sjenerøse grenser og pålitelig tilgang til åpen kildekode-modellene GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash.",
   "go.faq.q4": "Hva koster Go?",
   "go.faq.a4.p1.beforePricing": "Go koster",
   "go.faq.a4.p1.pricingLink": "$5 første måned",
@@ -350,7 +350,7 @@ export const dict = {
 
   "go.faq.q9": "Hva er forskjellen mellom gratis modeller og Go?",
   "go.faq.a9":
-    "Gratis modeller inkluderer Big Pickle pluss kampanjemodeller tilgjengelig på det tidspunktet, med en kvote på 200 forespørsler/dag. Go inkluderer GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash med høyere kvoter håndhevet over rullerende vinduer (5 timer, ukentlig og månedlig), omtrent tilsvarende $12 per 5 timer, $30 per uke og $60 per måned (faktiske forespørselsantall varierer etter modell og bruk).",
+    "Gratis modeller inkluderer Big Pickle pluss kampanjemodeller tilgjengelig på det tidspunktet, med en kvote på 200 forespørsler/dag. Go inkluderer GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro og DeepSeek V4 Flash med høyere kvoter håndhevet over rullerende vinduer (5 timer, ukentlig og månedlig), omtrent tilsvarende $12 per 5 timer, $30 per uke og $60 per måned (faktiske forespørselsantall varierer etter modell og bruk).",
 
   "zen.api.error.rateLimitExceeded": "Rate limit overskredet. Vennligst prøv igjen senere.",
   "zen.api.error.modelNotSupported": "Modell {{model}} støttes ikke",

packages/console/app/src/i18n/pl.ts

@@ -252,7 +252,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Niskokosztowe modele do kodowania dla każdego",
   "go.meta.description":
-    "Go zaczyna się od $5 za pierwszy miesiąc, potem $10/miesiąc, z hojnymi 5-godzinnymi limitami zapytań dla GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro i DeepSeek V4 Flash.",
+    "Go zaczyna się od $5 za pierwszy miesiąc, potem $10/miesiąc, z hojnymi 5-godzinnymi limitami zapytań dla GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro i DeepSeek V4 Flash.",
   "go.hero.title": "Niskokosztowe modele do kodowania dla każdego",
   "go.hero.body":
     "Go udostępnia programowanie z agentami programistom na całym świecie. Oferuje hojne limity i niezawodny dostęp do najzdolniejszych modeli open source, dzięki czemu możesz budować za pomocą potężnych agentów, nie martwiąc się o koszty czy dostępność.",
@@ -301,7 +301,7 @@ export const dict = {
   "go.problem.item2": "Hojne limity i niezawodny dostęp",
   "go.problem.item3": "Stworzony dla jak największej liczby programistów",
   "go.problem.item4":
-    "Zawiera GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro i DeepSeek V4 Flash",
+    "Zawiera GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro i DeepSeek V4 Flash",
   "go.how.title": "Jak działa Go",
   "go.how.body":
     "Go zaczyna się od $5 za pierwszy miesiąc, potem $10/miesiąc. Możesz go używać z OpenCode lub dowolnym agentem.",
@@ -327,7 +327,7 @@ export const dict = {
   "go.faq.a2": "Go obejmuje poniższe modele z wysokimi limitami i niezawodnym dostępem.",
   "go.faq.q3": "Czy Go to to samo co Zen?",
   "go.faq.a3":
-    "Nie. Zen to model płatności za użycie, podczas gdy Go zaczyna się od $5 za pierwszy miesiąc, potem $10/miesiąc, z hojnymi limitami i niezawodnym dostępem do modeli open source GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro i DeepSeek V4 Flash.",
+    "Nie. Zen to model płatności za użycie, podczas gdy Go zaczyna się od $5 za pierwszy miesiąc, potem $10/miesiąc, z hojnymi limitami i niezawodnym dostępem do modeli open source GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro i DeepSeek V4 Flash.",
   "go.faq.q4": "Ile kosztuje Go?",
   "go.faq.a4.p1.beforePricing": "Go kosztuje",
   "go.faq.a4.p1.pricingLink": "$5 za pierwszy miesiąc",
@@ -351,7 +351,7 @@ export const dict = {
 
   "go.faq.q9": "Jaka jest różnica między darmowymi modelami a Go?",
   "go.faq.a9":
-    "Darmowe modele obejmują Big Pickle oraz modele promocyjne dostępne w danym momencie, z limitem 200 zapytań/dzień. Go zawiera GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro i DeepSeek V4 Flash z wyższymi limitami zapytań egzekwowanymi w oknach kroczących (5-godzinnych, tygodniowych i miesięcznych), w przybliżeniu równoważnymi $12 na 5 godzin, $30 tygodniowo i $60 miesięcznie (rzeczywista liczba zapytań zależy od modelu i użycia).",
+    "Darmowe modele obejmują Big Pickle oraz modele promocyjne dostępne w danym momencie, z limitem 200 zapytań/dzień. Go zawiera GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro i DeepSeek V4 Flash z wyższymi limitami zapytań egzekwowanymi w oknach kroczących (5-godzinnych, tygodniowych i miesięcznych), w przybliżeniu równoważnymi $12 na 5 godzin, $30 tygodniowo i $60 miesięcznie (rzeczywista liczba zapytań zależy od modelu i użycia).",
 
   "zen.api.error.rateLimitExceeded": "Przekroczono limit zapytań. Spróbuj ponownie później.",
   "zen.api.error.modelNotSupported": "Model {{model}} nie jest obsługiwany",

packages/console/app/src/i18n/ru.ts

@@ -255,7 +255,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Недорогие модели для кодинга для всех",
   "go.meta.description":
-    "Go начинается с $5 за первый месяц, затем $10/месяц, с щедрыми лимитами запросов за 5 часов для GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro и DeepSeek V4 Flash.",
+    "Go начинается с $5 за первый месяц, затем $10/месяц, с щедрыми лимитами запросов за 5 часов для GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro и DeepSeek V4 Flash.",
   "go.hero.title": "Недорогие модели для кодинга для всех",
   "go.hero.body":
     "Go открывает доступ к агентам-программистам разработчикам по всему миру. Предлагая щедрые лимиты и надежный доступ к наиболее способным моделям с открытым исходным кодом, вы можете создавать проекты с мощными агентами, не беспокоясь о затратах или доступности.",
@@ -305,7 +305,7 @@ export const dict = {
   "go.problem.item2": "Щедрые лимиты и надежный доступ",
   "go.problem.item3": "Создан для максимального числа программистов",
   "go.problem.item4":
-    "Включает GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro и DeepSeek V4 Flash",
+    "Включает GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro и DeepSeek V4 Flash",
   "go.how.title": "Как работает Go",
   "go.how.body":
     "Go начинается с $5 за первый месяц, затем $10/месяц. Вы можете использовать его с OpenCode или любым агентом.",
@@ -331,7 +331,7 @@ export const dict = {
   "go.faq.a2": "Go включает перечисленные ниже модели с щедрыми лимитами и надежным доступом.",
   "go.faq.q3": "Go — это то же самое, что и Zen?",
   "go.faq.a3":
-    "Нет. Zen - это оплата по мере использования, в то время как Go начинается с $5 за первый месяц, затем $10/месяц, с щедрыми лимитами и надежным доступом к моделям с открытым исходным кодом GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro и DeepSeek V4 Flash.",
+    "Нет. Zen - это оплата по мере использования, в то время как Go начинается с $5 за первый месяц, затем $10/месяц, с щедрыми лимитами и надежным доступом к моделям с открытым исходным кодом GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro и DeepSeek V4 Flash.",
   "go.faq.q4": "Сколько стоит Go?",
   "go.faq.a4.p1.beforePricing": "Go стоит",
   "go.faq.a4.p1.pricingLink": "$5 за первый месяц",
@@ -355,7 +355,7 @@ export const dict = {
 
   "go.faq.q9": "В чем разница между бесплатными моделями и Go?",
   "go.faq.a9":
-    "Бесплатные модели включают Big Pickle плюс промо-модели, доступные на данный момент, с квотой 200 запросов/день. Go включает GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro и DeepSeek V4 Flash с более высокими квотами запросов, применяемыми в скользящих окнах (5 часов, неделя и месяц), что примерно эквивалентно $12 за 5 часов, $30 в неделю и $60 в месяц (фактическое количество запросов зависит от модели и использования).",
+    "Бесплатные модели включают Big Pickle плюс промо-модели, доступные на данный момент, с квотой 200 запросов/день. Go включает GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro и DeepSeek V4 Flash с более высокими квотами запросов, применяемыми в скользящих окнах (5 часов, неделя и месяц), что примерно эквивалентно $12 за 5 часов, $30 в неделю и $60 в месяц (фактическое количество запросов зависит от модели и использования).",
 
   "zen.api.error.rateLimitExceeded": "Превышен лимит запросов. Пожалуйста, попробуйте позже.",
   "zen.api.error.modelNotSupported": "Модель {{model}} не поддерживается",

packages/console/app/src/i18n/th.ts

@@ -250,7 +250,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | โมเดลเขียนโค้ดราคาประหยัดสำหรับทุกคน",
   "go.meta.description":
-    "Go เริ่มต้นที่ $5 สำหรับเดือนแรก จากนั้น $10/เดือน พร้อมขีดจำกัดคำขอ 5 ชั่วโมงที่เอื้อเฟื้อสำหรับ GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro และ DeepSeek V4 Flash",
+    "Go เริ่มต้นที่ $5 สำหรับเดือนแรก จากนั้น $10/เดือน พร้อมขีดจำกัดคำขอ 5 ชั่วโมงที่เอื้อเฟื้อสำหรับ GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro และ DeepSeek V4 Flash",
   "go.hero.title": "โมเดลเขียนโค้ดราคาประหยัดสำหรับทุกคน",
   "go.hero.body":
     "Go นำการเขียนโค้ดแบบเอเจนต์มาสู่นักเขียนโปรแกรมทั่วโลก เสนอขีดจำกัดที่กว้างขวางและการเข้าถึงโมเดลโอเพนซอร์สที่มีความสามารถสูงสุดได้อย่างน่าเชื่อถือ เพื่อให้คุณสามารถสร้างสรรค์ด้วยเอเจนต์ที่ทรงพลังโดยไม่ต้องกังวลเรื่องค่าใช้จ่ายหรือความพร้อมใช้งาน",
@@ -298,7 +298,7 @@ export const dict = {
   "go.problem.item2": "ขีดจำกัดที่กว้างขวางและการเข้าถึงที่เชื่อถือได้",
   "go.problem.item3": "สร้างขึ้นเพื่อโปรแกรมเมอร์จำนวนมากที่สุดเท่าที่จะเป็นไปได้",
   "go.problem.item4":
-    "รวมถึง GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro และ DeepSeek V4 Flash",
+    "รวมถึง GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro และ DeepSeek V4 Flash",
   "go.how.title": "Go ทำงานอย่างไร",
   "go.how.body": "Go เริ่มต้นที่ $5 สำหรับเดือนแรก จากนั้น $10/เดือน คุณสามารถใช้กับ OpenCode หรือเอเจนต์ใดก็ได้",
   "go.how.step1.title": "สร้างบัญชี",
@@ -323,7 +323,7 @@ export const dict = {
   "go.faq.a2": "Go รวมโมเดลด้านล่างนี้ พร้อมขีดจำกัดที่มากและการเข้าถึงที่เชื่อถือได้",
   "go.faq.q3": "Go เหมือนกับ Zen หรือไม่?",
   "go.faq.a3":
-    "ไม่ Zen เป็นแบบจ่ายตามการใช้งาน ในขณะที่ Go เริ่มต้นที่ $5 สำหรับเดือนแรก จากนั้น $10/เดือน พร้อมขีดจำกัดที่เอื้อเฟื้อและการเข้าถึงโมเดลโอเพนซอร์ส GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro และ DeepSeek V4 Flash อย่างเชื่อถือได้",
+    "ไม่ Zen เป็นแบบจ่ายตามการใช้งาน ในขณะที่ Go เริ่มต้นที่ $5 สำหรับเดือนแรก จากนั้น $10/เดือน พร้อมขีดจำกัดที่เอื้อเฟื้อและการเข้าถึงโมเดลโอเพนซอร์ส GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro และ DeepSeek V4 Flash อย่างเชื่อถือได้",
   "go.faq.q4": "Go ราคาเท่าไหร่?",
   "go.faq.a4.p1.beforePricing": "Go ราคา",
   "go.faq.a4.p1.pricingLink": "$5 เดือนแรก",
@@ -346,7 +346,7 @@ export const dict = {
 
   "go.faq.q9": "ความแตกต่างระหว่างโมเดลฟรีและ Go คืออะไร?",
   "go.faq.a9":
-    "โมเดลฟรีรวมถึง Big Pickle บวกกับโมเดลโปรโมชั่นที่มีให้ในขณะนั้น ด้วยโควต้า 200 คำขอ/วัน Go รวมถึง GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro และ DeepSeek V4 Flash ที่มีโควต้าคำขอสูงกว่า ซึ่งบังคับใช้ผ่านช่วงเวลาหมุนเวียน (5 ชั่วโมง, รายสัปดาห์ และรายเดือน) เทียบเท่าประมาณ $12 ต่อ 5 ชั่วโมง, $30 ต่อสัปดาห์ และ $60 ต่อเดือน (จำนวนคำขอจริงจะแตกต่างกันไปตามโมเดลและการใช้งาน)",
+    "โมเดลฟรีรวมถึง Big Pickle บวกกับโมเดลโปรโมชั่นที่มีให้ในขณะนั้น ด้วยโควต้า 200 คำขอ/วัน Go รวมถึง GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro และ DeepSeek V4 Flash ที่มีโควต้าคำขอสูงกว่า ซึ่งบังคับใช้ผ่านช่วงเวลาหมุนเวียน (5 ชั่วโมง, รายสัปดาห์ และรายเดือน) เทียบเท่าประมาณ $12 ต่อ 5 ชั่วโมง, $30 ต่อสัปดาห์ และ $60 ต่อเดือน (จำนวนคำขอจริงจะแตกต่างกันไปตามโมเดลและการใช้งาน)",
 
   "zen.api.error.rateLimitExceeded": "เกินขีดจำกัดอัตราการใช้งาน กรุณาลองใหม่ในภายหลัง",
   "zen.api.error.modelNotSupported": "ไม่รองรับโมเดล {{model}}",

packages/console/app/src/i18n/tr.ts

@@ -253,7 +253,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Herkes için düşük maliyetli kodlama modelleri",
   "go.meta.description":
-    "Go ilk ay $5, sonrasında ayda 10$ fiyatıyla başlar; GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro ve DeepSeek V4 Flash için cömert 5 saatlik istek limitleri sunar.",
+    "Go ilk ay $5, sonrasında ayda 10$ fiyatıyla başlar; GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro ve DeepSeek V4 Flash için cömert 5 saatlik istek limitleri sunar.",
   "go.hero.title": "Herkes için düşük maliyetli kodlama modelleri",
   "go.hero.body":
     "Go, dünya çapındaki programcılara ajan tabanlı kodlama getiriyor. En yetenekli açık kaynaklı modellere cömert limitler ve güvenilir erişim sunarak, maliyet veya erişilebilirlik konusunda endişelenmeden güçlü ajanlarla geliştirme yapmanızı sağlar.",
@@ -303,7 +303,7 @@ export const dict = {
   "go.problem.item2": "Cömert limitler ve güvenilir erişim",
   "go.problem.item3": "Mümkün olduğunca çok programcı için geliştirildi",
   "go.problem.item4":
-    "GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro ve DeepSeek V4 Flash içerir",
+    "GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro ve DeepSeek V4 Flash içerir",
   "go.how.title": "Go nasıl çalışır?",
   "go.how.body":
     "Go ilk ay $5, sonrasında ayda 10$ fiyatıyla başlar. OpenCode veya herhangi bir ajanla kullanabilirsiniz.",
@@ -329,7 +329,7 @@ export const dict = {
   "go.faq.a2": "Go, aşağıda listelenen modelleri cömert limitler ve güvenilir erişimle sunar.",
   "go.faq.q3": "Go, Zen ile aynı mı?",
   "go.faq.a3":
-    "Hayır. Zen kullandıkça öde modelidir, Go ise ilk ay $5, sonrasında ayda 10$ fiyatıyla başlar; GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro ve DeepSeek V4 Flash açık kaynak modellerine cömert limitler ve güvenilir erişim sunar.",
+    "Hayır. Zen kullandıkça öde modelidir, Go ise ilk ay $5, sonrasında ayda 10$ fiyatıyla başlar; GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro ve DeepSeek V4 Flash açık kaynak modellerine cömert limitler ve güvenilir erişim sunar.",
   "go.faq.q4": "Go ne kadar?",
   "go.faq.a4.p1.beforePricing": "Go'nun maliyeti",
   "go.faq.a4.p1.pricingLink": "İlk ay $5",
@@ -353,7 +353,7 @@ export const dict = {
 
   "go.faq.q9": "Ücretsiz modeller ve Go arasındaki fark nedir?",
   "go.faq.a9":
-    "Ücretsiz modeller, günlük 200 istek kotası ile Big Pickle ve o sırada mevcut olan promosyonel modelleri içerir. Go ise GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro ve DeepSeek V4 Flash modellerini; yuvarlanan pencereler (5 saatlik, haftalık ve aylık) üzerinden uygulanan daha yüksek istek kotalarıyla içerir. Bu kotalar kabaca her 5 saatte 12$, haftada 30$ ve ayda 60$ değerine eşdeğerdir (gerçek istek sayıları modele ve kullanıma göre değişir).",
+    "Ücretsiz modeller, günlük 200 istek kotası ile Big Pickle ve o sırada mevcut olan promosyonel modelleri içerir. Go ise GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro ve DeepSeek V4 Flash modellerini; yuvarlanan pencereler (5 saatlik, haftalık ve aylık) üzerinden uygulanan daha yüksek istek kotalarıyla içerir. Bu kotalar kabaca her 5 saatte 12$, haftada 30$ ve ayda 60$ değerine eşdeğerdir (gerçek istek sayıları modele ve kullanıma göre değişir).",
 
   "zen.api.error.rateLimitExceeded": "İstek limiti aşıldı. Lütfen daha sonra tekrar deneyin.",
   "zen.api.error.modelNotSupported": "{{model}} modeli desteklenmiyor",

packages/console/app/src/i18n/uk.ts

@@ -252,7 +252,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | Недорогі моделі кодування для всіх",
   "go.meta.description":
-    "Go починається від $5 за перший місяць, потім $10/місяць, з generous 5-годинними лімітами запитів для GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro та DeepSeek V4 Flash.",
+    "Go починається від $5 за перший місяць, потім $10/місяць, з generous 5-годинними лімітами запитів для GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro та DeepSeek V4 Flash.",
   "go.hero.title": "Недорогі моделі кодування для всіх",
   "go.hero.body":
     "Go надає агентне програмування програмістам у всьому світі, пропонуючи щедрі ліміти та надійний доступ до найкращих моделей з відкритим кодом.",
@@ -301,7 +301,7 @@ export const dict = {
   "go.problem.item2": "Щедрі ліміти та надійний доступ",
   "go.problem.item3": "Створено для якомога більшої кількості програмістів",
   "go.problem.item4":
-    "Включає GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro та DeepSeek V4 Flash",
+    "Включає GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro та DeepSeek V4 Flash",
   "go.how.title": "Як працює Go",
   "go.how.body":
     "Go починається від $5 за перший місяць, потім $10/місяць. Використовуйте з OpenCode або будь-яким агентом.",
@@ -327,7 +327,7 @@ export const dict = {
   "go.faq.a2": "Go включає моделі, перелічені нижче, із щедрими лімітами та надійним доступом.",
   "go.faq.q3": "Чи Go те саме, що Zen?",
   "go.faq.a3":
-    "Ні. Zen — це плата за використання, тоді як Go починається від $5 за перший місяць, потім $10/місяць, із щедрими лімітами та надійним доступом до моделей з відкритим кодом GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro та DeepSeek V4 Flash.",
+    "Ні. Zen — це плата за використання, тоді як Go починається від $5 за перший місяць, потім $10/місяць, із щедрими лімітами та надійним доступом до моделей з відкритим кодом.",
   "go.faq.q4": "Скільки коштує Go?",
   "go.faq.a4.p1.beforePricing": "Go коштує",
   "go.faq.a4.p1.pricingLink": "$5 за перший місяць",
@@ -350,7 +350,7 @@ export const dict = {
 
   "go.faq.q9": "Яка різниця між безкоштовними моделями та Go?",
   "go.faq.a9":
-    "Безкоштовні моделі включають Big Pickle та акційні моделі з лімітом 200 запитів/день. Go включає GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro та DeepSeek V4 Flash із вищими лімітами.",
+    "Безкоштовні моделі включають Big Pickle та акційні моделі з лімітом 200 запитів/день. Go включає GLM-5.1, GLM-5, Kimi K2.5, Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro та DeepSeek V4 Flash із вищими лімітами.",
 
   "zen.api.error.rateLimitExceeded": "Перевищено ліміт запитів. Спробуйте пізніше.",
   "zen.api.error.modelNotSupported": "Модель {{model}} не підтримується",

packages/console/app/src/i18n/zh.ts

@@ -241,7 +241,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | 人人可用的低成本编程模型",
   "go.meta.description":
-    "Go 首月 $5，之后 $10/月，提供对 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash 的 5 小时充裕请求额度。",
+    "Go 首月 $5，之后 $10/月，提供对 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、Qwen3.5 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash 的 5 小时充裕请求额度。",
   "go.hero.title": "人人可用的低成本编程模型",
   "go.hero.body":
     "Go 将代理编程带给全世界的程序员。提供充裕的限额和对最强大的开源模型的可靠访问，让您可以利用强大的代理进行构建，而无需担心成本或可用性。",
@@ -289,7 +289,7 @@ export const dict = {
   "go.problem.item2": "充裕的限额和可靠的访问",
   "go.problem.item3": "为尽可能多的程序员打造",
   "go.problem.item4":
-    "包含 GLM-5.1, GLM-5, Kimi K2.5、Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash",
+    "包含 GLM-5.1, GLM-5, Kimi K2.5、Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash",
   "go.how.title": "Go 如何工作",
   "go.how.body": "Go 起价为首月 $5，之后 $10/月。您可以将其与 OpenCode 或任何代理搭配使用。",
   "go.how.step1.title": "创建账户",
@@ -311,7 +311,7 @@ export const dict = {
   "go.faq.a2": "Go 包含下方列出的模型，提供充足的限额和可靠的访问。",
   "go.faq.q3": "Go 和 Zen 一样吗？",
   "go.faq.a3":
-    "不。Zen 是按量付费，而 Go 首月 $5，之后 $10/月，提供充裕的额度，并可可靠地访问 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash 等开源模型。",
+    "不。Zen 是按量付费，而 Go 首月 $5，之后 $10/月，提供充裕的额度，并可可靠地访问 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、Qwen3.5 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash 等开源模型。",
   "go.faq.q4": "Go 多少钱？",
   "go.faq.a4.p1.beforePricing": "Go 费用为",
   "go.faq.a4.p1.pricingLink": "首月 $5",
@@ -333,7 +333,7 @@ export const dict = {
 
   "go.faq.q9": "免费模型和 Go 之间的区别是什么？",
   "go.faq.a9":
-    "免费模型包含 Big Pickle 加上当时可用的促销模型，每天有 200 次请求的配额。Go 包含 GLM-5.1, GLM-5, Kimi K2.5、Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash，并在滚动窗口（5 小时、每周和每月）内执行更高的请求配额，大致相当于每 5 小时 $12、每周 $30 和每月 $60（实际请求计数因模型和使用情况而异）。",
+    "免费模型包含 Big Pickle 加上当时可用的促销模型，每天有 200 次请求的配额。Go 包含 GLM-5.1, GLM-5, Kimi K2.5、Kimi K2.6, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.7 Max, Qwen3.6 Plus, Qwen3.5 Plus, MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash，并在滚动窗口（5 小时、每周和每月）内执行更高的请求配额，大致相当于每 5 小时 $12、每周 $30 和每月 $60（实际请求计数因模型和使用情况而异）。",
 
   "zen.api.error.rateLimitExceeded": "超出速率限制。请稍后重试。",
   "zen.api.error.modelNotSupported": "不支持模型 {{model}}",

packages/console/app/src/i18n/zht.ts

@@ -241,7 +241,7 @@ export const dict = {
 
   "go.title": "OpenCode Go | 低成本全民編碼模型",
   "go.meta.description":
-    "Go 首月 $5，之後 $10/月，提供對 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash 的 5 小時充裕請求額度。",
+    "Go 首月 $5，之後 $10/月，提供對 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、Qwen3.5 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash 的 5 小時充裕請求額度。",
   "go.hero.title": "低成本全民編碼模型",
   "go.hero.body":
     "Go 將代理編碼帶給全世界的程式設計師。提供寬裕的限額以及對最強大開源模型的穩定存取，讓你可以使用強大的代理進行構建，而無需擔心成本或可用性。",
@@ -289,7 +289,7 @@ export const dict = {
   "go.problem.item2": "寬裕的限額與穩定存取",
   "go.problem.item3": "專為盡可能多的程式設計師打造",
   "go.problem.item4":
-    "包含 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 與 DeepSeek V4 Flash",
+    "包含 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、Qwen3.5 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 與 DeepSeek V4 Flash",
   "go.how.title": "Go 如何運作",
   "go.how.body": "Go 起價為首月 $5，之後 $10/月。您可以將其與 OpenCode 或任何代理搭配使用。",
   "go.how.step1.title": "建立帳號",
@@ -311,7 +311,7 @@ export const dict = {
   "go.faq.a2": "Go 包含下方列出的模型，提供充足的額度與穩定的存取。",
   "go.faq.q3": "Go 與 Zen 一樣嗎？",
   "go.faq.a3":
-    "不。Zen 是按量付費，而 Go 首月 $5，之後 $10/月，提供充裕的額度，並可可靠地存取 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash 等開源模型。",
+    "不。Zen 是按量付費，而 Go 首月 $5，之後 $10/月，提供充裕的額度，並可可靠地存取 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、Qwen3.5 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 和 DeepSeek V4 Flash 等開源模型。",
   "go.faq.q4": "Go 費用是多少？",
   "go.faq.a4.p1.beforePricing": "Go 費用為",
   "go.faq.a4.p1.pricingLink": "首月 $5",
@@ -333,7 +333,7 @@ export const dict = {
 
   "go.faq.q9": "免費模型與 Go 有什麼區別？",
   "go.faq.a9":
-    "免費模型包括 Big Pickle 以及當時可用的促銷模型，配額為 200 次請求/天。Go 包括 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 與 DeepSeek V4 Flash，並在滾動視窗（5 小時、每週和每月）內執行更高的請求配額，大約相當於每 5 小時 $12、每週 $30 和每月 $60（實際請求數因模型和使用情況而異）。",
+    "免費模型包括 Big Pickle 以及當時可用的促銷模型，配額為 200 次請求/天。Go 包括 GLM-5.1、GLM-5、Kimi K2.5、Kimi K2.6、MiMo-V2.5-Pro、MiMo-V2.5、Qwen3.7 Max、Qwen3.6 Plus、Qwen3.5 Plus、MiniMax M2.5、MiniMax M2.7、DeepSeek V4 Pro 與 DeepSeek V4 Flash，並在滾動視窗（5 小時、每週和每月）內執行更高的請求配額，大約相當於每 5 小時 $12、每週 $30 和每月 $60（實際請求數因模型和使用情況而異）。",
 
   "zen.api.error.rateLimitExceeded": "超出頻率限制。請稍後再試。",
   "zen.api.error.modelNotSupported": "不支援模型 {{model}}",

packages/console/app/src/lib/referral-invite.ts

@@ -1,6 +1,4 @@
-import { Actor } from "@opencode-ai/console-core/actor.js"
 import { Referral } from "@opencode-ai/console-core/referral.js"
-import { getRequestEvent } from "solid-js/web"
 
 const REFERRAL_COOKIE = "oc_referral"
 const REFERRAL_MAX_AGE = 60 * 60 * 24 * 30
@@ -28,17 +26,3 @@ export function referralCodeFromCookieHeader(header: string | null) {
       ?.slice(`${REFERRAL_COOKIE}=`.length),
   )
 }
-
-export async function createReferralFromCookie() {
-  const event = getRequestEvent()
-  const referralCode = referralCodeFromCookieHeader(event?.request.headers.get("cookie") ?? null)
-  if (!referralCode) return
-
-  await Referral.createFromAccount({
-    accountID: Actor.account(),
-    referralCode,
-  }).catch((error) => {
-    console.error("Referral create failed", error)
-  })
-  event?.response.headers.append("set-cookie", clearReferralCookie())
-}

packages/console/app/src/lib/stats-proxy.ts

@@ -1,37 +0,0 @@
-import type { APIEvent } from "@solidjs/start/server"
-import { Resource } from "@opencode-ai/console-resource"
-
-export async function statsProxy(evt: APIEvent) {
-  const req = evt.request.clone()
-  const targetUrl = new URL(req.url)
-  targetUrl.protocol = "https:"
-  targetUrl.hostname = Resource.App.stage === "production" ? "stats.opencode.ai" : "stats.dev.opencode.ai"
-  targetUrl.port = ""
-
-  if (targetUrl.pathname.startsWith("/stats/_build/")) {
-    targetUrl.pathname = targetUrl.pathname.slice("/stats".length)
-  }
-
-  const response = await fetch(targetUrl, {
-    method: req.method,
-    headers: req.headers,
-    body: req.body,
-  })
-
-  if (!response.headers.get("content-type")?.includes("text/html")) return response
-
-  const headers = new Headers(response.headers)
-  headers.delete("content-encoding")
-  headers.delete("content-length")
-  headers.delete("etag")
-
-  return new Response(rewriteStatsHtml(await response.text()), {
-    status: response.status,
-    statusText: response.statusText,
-    headers,
-  })
-}
-
-function rewriteStatsHtml(html: string) {
-  return html.replaceAll('"/_build/', '"/stats/_build/').replaceAll("'/_build/", "'/stats/_build/")
-}

packages/console/app/src/routes/go/index.css

@@ -90,7 +90,6 @@ body {
 
 [data-page="go"] {
   background: var(--color-background);
-  overflow-x: clip;
   --padding: 5rem;
   --vertical-padding: 4rem;
 

packages/console/app/src/routes/go/index.tsx

@@ -31,6 +31,7 @@ const models = [
   { name: "MiMo-V2.5", provider: "Xiaomi MiMo" },
   { name: "Qwen3.7 Max", provider: "Alibaba Cloud Model Studio" },
   { name: "Qwen3.6 Plus", provider: "Alibaba Cloud Model Studio" },
+  { name: "Qwen3.5 Plus", provider: "Alibaba Cloud Model Studio" },
   { name: "MiniMax M2.7", provider: "MiniMax" },
   { name: "MiniMax M2.5", provider: "MiniMax" },
   { name: "DeepSeek V4 Pro", provider: "DeepSeek" },
@@ -67,6 +68,7 @@ function LimitsGraph(props: { href: string }) {
     { id: "qwen3.6-plus", name: "Qwen3.6 Plus", req: 3300, d: "280ms" },
     { id: "minimax-m2.7", name: "MiniMax M2.7", req: 3400, d: "300ms" },
     { id: "deepseek-v4-pro", name: "DeepSeek V4 Pro", req: 3450, d: "200ms" },
+    { id: "qwen3.5-plus", name: "Qwen3.5 Plus", req: 10200, d: "360ms" },
     { id: "mimo-v2.5", name: "MiMo-V2.5", req: 30100, d: "340ms" },
     { id: "deepseek-v4-flash", name: "DeepSeek V4 Flash", req: 31650, d: "340ms" },
   ]

packages/console/app/src/routes/stats/[...path].ts

@@ -1,8 +0,0 @@
-import { statsProxy } from "~/lib/stats-proxy"
-
-export const GET = statsProxy
-export const POST = statsProxy
-export const PUT = statsProxy
-export const DELETE = statsProxy
-export const OPTIONS = statsProxy
-export const PATCH = statsProxy

packages/console/app/src/routes/stats/index.ts

@@ -1,8 +0,0 @@
-import { statsProxy } from "~/lib/stats-proxy"
-
-export const GET = statsProxy
-export const POST = statsProxy
-export const PUT = statsProxy
-export const DELETE = statsProxy
-export const OPTIONS = statsProxy
-export const PATCH = statsProxy

packages/console/app/src/routes/workspace/[id]/go/lite-section.tsx

@@ -15,7 +15,6 @@ import { useI18n } from "~/context/i18n"
 import { useLanguage } from "~/context/language"
 import { formError } from "~/lib/form-error"
 import { formatResetTime, liteResetTimeKeys } from "~/lib/format-reset-time"
-import { createReferralFromCookie } from "~/lib/referral-invite"
 
 import { IconAlipay, IconUpi } from "~/component/icon"
 
@@ -75,14 +74,16 @@ const createLiteCheckoutUrl = action(
   async (workspaceID: string, successUrl: string, cancelUrl: string, method?: "alipay" | "upi") => {
     "use server"
     return json(
-      await withActor(async () => {
-        const data = await Billing.generateLiteCheckoutUrl({ successUrl, cancelUrl, method })
-        await createReferralFromCookie()
-        return { error: undefined, data }
-      }, workspaceID).catch((e) => ({
-        error: e.message as string,
-        data: undefined,
-      })),
+      await withActor(
+        () =>
+          Billing.generateLiteCheckoutUrl({ successUrl, cancelUrl, method })
+            .then((data) => ({ error: undefined, data }))
+            .catch((e) => ({
+              error: e.message as string,
+              data: undefined,
+            })),
+        workspaceID,
+      ),
       { revalidate: [queryBillingInfo.key, queryLiteSubscription.key] },
     )
   },
@@ -265,6 +266,7 @@ export function LiteSection(props: { lite: LiteSubscription | undefined }) {
             <li>MiMo-V2.5</li>
             <li>MiniMax M2.5</li>
             <li>MiniMax M2.7</li>
+            <li>Qwen3.5 Plus</li>
             <li>Qwen3.6 Plus</li>
             <li>Qwen3.7 Max</li>
             <li>DeepSeek V4 Pro</li>

packages/console/app/src/routes/zen/util/handler.ts

@@ -47,7 +47,6 @@ import { i18n, type Key } from "~/i18n"
 import { localeFromRequest } from "~/lib/language"
 import { createModelTpmLimiter } from "./modelTpmLimiter"
 import { createModelTpsLimiter } from "./modelTpsLimiter"
-import { accumulateUsage, HOT_WORKSPACES } from "./usageBatcher"
 
 type ZenData = Awaited<ReturnType<typeof ZenData.list>>
 type RetryOptions = {
@@ -479,24 +478,29 @@ export async function handler(
     stickyId: string,
     trialProviders: string[] | undefined,
     retry: RetryOptions,
-    stickyProviderId: string | undefined,
+    stickyProvider: string | undefined,
     modelTpmLimits: Record<string, number> | undefined,
-    modelTpsLimits: Record<string, { qualify: number; unqualify: number }> | undefined,
+    modelTpsLimits: Record<string, boolean> | undefined,
   ) {
     const modelProvider = (() => {
+      const allProviders = modelInfo.providers.filter((provider) => !provider.disabled)
+
       // Byok is top priority b/c if user set their own API key, we should use it
       // instead of using the sticky provider for the same session
       if (authInfo?.provider?.credentials) {
-        return modelInfo.providers.find((provider) => provider.id === modelInfo.byokProvider)
+        return allProviders.find((provider) => provider.id === modelInfo.byokProvider)
+      }
+
+      // Always use the same provider for the same session
+      if (stickyProvider) {
+        const provider = allProviders.find((provider) => provider.id === stickyProvider)
+        if (provider) return provider
       }
 
-      // Prioritize trial providers
-      let allProviders = modelInfo.providers.filter((provider) => !provider.disabled)
       if (trialProviders) {
-        allProviders = allProviders.map((provider) => ({
-          ...provider,
-          priority: trialProviders.includes(provider.id) ? 0 : provider.priority,
-        }))
+        const trialProvider = trialProviders[Math.floor(Math.random() * trialProviders.length)]
+        const provider = allProviders.find((provider) => provider.id === trialProvider)
+        if (provider) return provider
       }
 
       if (retry.retryCount !== MAX_FAILOVER_RETRIES) {
@@ -511,11 +515,7 @@ export async function handler(
           })
           .filter((provider) => {
             if (!provider.tpsGoal) return true
-            const tps = modelTpsLimits?.[`${provider.id}/${provider.model}/${provider.tpsGoal}`] ?? {
-              qualify: 0,
-              unqualify: 0,
-            }
-            const isLowTps = tps.qualify + tps.unqualify > 10 && tps.qualify < tps.unqualify
+            const isLowTps = modelTpsLimits?.[`${provider.id}/${provider.model}/${provider.tpsGoal}`] ?? false
             return !isLowTps
           })
           .map((provider) => {
@@ -533,23 +533,7 @@ export async function handler(
         }
         const index = (h >>> 0) % providers.length // make unsigned + range 0..length-1
         const provider = providers[index || 0]
-
-        // sticky provider does not exist => use selected provider
-        if (!stickyProviderId) return provider
-        const stickProvider = allProviders.find((provider) => provider.id === stickyProviderId)
-        if (!stickProvider) return provider
-
-        // stick provider exists + selected provider is API type => use sticky provider
-        if (!provider.tpsGoal) return stickProvider
-
-        // stick provier exists + selected provider is GPU type + GPU not idle => use selected provider
-        const tps = modelTpsLimits?.[`${provider.id}/${provider.model}/${provider.tpsGoal}`] ?? {
-          qualify: 0,
-          unqualify: 0,
-        }
-        if (tps.qualify <= tps.unqualify * 3) return stickProvider
-
-        return provider
+        if (provider) return provider
       }
 
       // fallback provider
@@ -982,19 +966,6 @@ export async function handler(
     authInfo = authInfo!
 
     const cost = centsToMicroCents(totalCostInCent)
-
-    // For hot workspaces, batch balance/usage updates through Redis to avoid
-    // row-level lock contention on BillingTable/UserTable. Returns the amount
-    // to flush this request, or null to skip the DB writes entirely.
-    const balanceFlush = await (async () => {
-      if (billingSource !== "subscription" && billingSource !== "lite" && HOT_WORKSPACES.has(authInfo.workspaceID)) {
-        const workspaceCost = billingSource === "free" || billingSource === "byok" ? 0 : cost
-        const flush = await accumulateUsage(authInfo.workspaceID, authInfo.user.id, workspaceCost, cost)
-        return { batched: true as const, flush }
-      }
-      return { batched: false as const, flush: null }
-    })()
-
     await Database.use((db) =>
       Promise.all([
         db.insert(UsageTable).values({
@@ -1096,22 +1067,18 @@ export async function handler(
             ]
           }
 
-          // Batched hot workspace: skip DB writes unless this request is the flush.
-          if (balanceFlush.batched && !balanceFlush.flush) return []
-
-          const workspaceDelta = balanceFlush.flush?.workspaceCost ?? cost
-          const userDelta = balanceFlush.flush?.userCost ?? cost
-          const balanceDelta = billingSource === "free" || billingSource === "byok" ? 0 : workspaceDelta
-
           return [
             db
               .update(BillingTable)
               .set({
-                balance: sql`${BillingTable.balance} - ${balanceDelta}`,
+                balance:
+                  billingSource === "free" || billingSource === "byok"
+                    ? sql`${BillingTable.balance} - ${0}`
+                    : sql`${BillingTable.balance} - ${cost}`,
                 monthlyUsage: sql`
               CASE
-                WHEN MONTH(${BillingTable.timeMonthlyUsageUpdated}) = MONTH(now()) AND YEAR(${BillingTable.timeMonthlyUsageUpdated}) = YEAR(now()) THEN ${BillingTable.monthlyUsage} + ${workspaceDelta}
-                ELSE ${workspaceDelta}
+                WHEN MONTH(${BillingTable.timeMonthlyUsageUpdated}) = MONTH(now()) AND YEAR(${BillingTable.timeMonthlyUsageUpdated}) = YEAR(now()) THEN ${BillingTable.monthlyUsage} + ${cost}
+                ELSE ${cost}
               END
             `,
                 timeMonthlyUsageUpdated: sql`now()`,
@@ -1122,8 +1089,8 @@ export async function handler(
               .set({
                 monthlyUsage: sql`
               CASE
-                WHEN MONTH(${UserTable.timeMonthlyUsageUpdated}) = MONTH(now()) AND YEAR(${UserTable.timeMonthlyUsageUpdated}) = YEAR(now()) THEN ${UserTable.monthlyUsage} + ${userDelta}
-                ELSE ${userDelta}
+                WHEN MONTH(${UserTable.timeMonthlyUsageUpdated}) = MONTH(now()) AND YEAR(${UserTable.timeMonthlyUsageUpdated}) = YEAR(now()) THEN ${UserTable.monthlyUsage} + ${cost}
+                ELSE ${cost}
               END
             `,
                 timeMonthlyUsageUpdated: sql`now()`,

packages/console/app/src/routes/zen/util/modelTpsLimiter.ts

@@ -37,7 +37,7 @@ export function createModelTpsLimiter(providers: { id: string; model: string; tp
       )
 
       // convert to map of model to summed count across current and previous intervals
-      return data.reduce(
+      const result = data.reduce(
         (acc, curr) => {
           const existing = acc[curr.id] ?? { qualify: 0, unqualify: 0 }
           acc[curr.id] = {
@@ -48,6 +48,13 @@ export function createModelTpsLimiter(providers: { id: string; model: string; tp
         },
         {} as Record<string, { qualify: number; unqualify: number }>,
       )
+
+      return Object.fromEntries(
+        Object.entries(result).map(([id, { qualify, unqualify }]) => {
+          const isLowTps = qualify + unqualify > 10 && qualify < unqualify
+          return [id, isLowTps]
+        }),
+      )
     },
     track: async (
       provider: string,

packages/console/app/src/routes/zen/util/usageBatcher.ts

@@ -1,31 +0,0 @@
-import { Resource } from "@opencode-ai/console-resource"
-import { getRedis } from "./redis"
-
-// Workspaces whose balance/usage updates should be batched in Redis to avoid
-// row-level lock contention on BillingTable / UserTable.
-export const HOT_WORKSPACES = new Set<string>([
-  "wrk_01KJ8PX5CH50Y4YNGNS9ZR8YDC", // invoice
-])
-
-// Probability that a given request flushes the accumulated totals to the DB.
-// Lower = fewer DB writes, more staleness. ~1 in 100 -> ~1% of requests write.
-const FLUSH_PROBABILITY = 1 / 100
-
-export async function accumulateUsage(workspaceID: string, userID: string, workspaceCost: number, userCost: number) {
-  const redis = getRedis()
-  const wKey = `${Resource.App.stage}:usage:wrk:${workspaceID}`
-  const uKey = `${Resource.App.stage}:usage:usr:${workspaceID}:${userID}`
-
-  await Promise.all([redis.incrby(wKey, workspaceCost), redis.incrby(uKey, userCost)])
-
-  if (Math.random() > FLUSH_PROBABILITY) return null
-
-  // Atomically take the current totals and reset to 0
-  const [workspaceTotal, userTotal] = await Promise.all([redis.getdel<number>(wKey), redis.getdel<number>(uKey)])
-
-  const workspaceFlush = Number(workspaceTotal ?? 0)
-  const userFlush = Number(userTotal ?? 0)
-  if (workspaceFlush === 0 && userFlush === 0) return null
-
-  return { workspaceCost: workspaceFlush, userCost: userFlush }
-}

packages/console/core/package.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://json.schemastore.org/package.json",
   "name": "@opencode-ai/console-core",
-  "version": "1.15.13",
+  "version": "1.15.11",
   "private": true,
   "type": "module",
   "license": "MIT",
@@ -37,7 +37,6 @@
     "update-limits": "script/update-limits.ts",
     "promote-limits-to-dev": "script/promote-limits.ts dev",
     "promote-limits-to-prod": "script/promote-limits.ts production",
-    "referral-backfill": "script/referral-backfill.ts",
     "typecheck": "tsgo --noEmit"
   },
   "devDependencies": {

packages/console/core/script/referral-backfill.ts

@@ -1,153 +0,0 @@
-import { and, Database, eq, inArray, isNull } from "../src/drizzle/index.js"
-import { Identifier } from "../src/identifier.js"
-import { Referral } from "../src/referral.js"
-import { LiteTable } from "../src/schema/billing.sql.js"
-import { ReferralRewardTable, ReferralTable } from "../src/schema/referral.sql.js"
-import { UserTable } from "../src/schema/user.sql.js"
-import { WorkspaceTable } from "../src/schema/workspace.sql.js"
-
-const backfills = [
-  {
-    inviterWorkspaceID: "wrk_00000000000000000000000000",
-    inviteeWorkspaceID: "wrk_00000000000000000000000000",
-    inviteeAccountID: "acc_00000000000000000000000000",
-  },
-]
-
-console.log(`Backfilling ${backfills.length} referrals`)
-
-for (const [index, backfill] of backfills.entries()) {
-  console.log(`[${index + 1}/${backfills.length}] ${backfill.inviterWorkspaceID} -> ${backfill.inviteeWorkspaceID}`)
-  console.log(`  invitee account: ${backfill.inviteeAccountID}`)
-
-  const result = await Database.transaction(async (tx) => {
-    if (backfill.inviterWorkspaceID === backfill.inviteeWorkspaceID) throw new Error("Self-referral workspace mismatch")
-
-    const inviterWorkspace = await tx
-      .select({ id: WorkspaceTable.id })
-      .from(WorkspaceTable)
-      .where(and(eq(WorkspaceTable.id, backfill.inviterWorkspaceID), isNull(WorkspaceTable.timeDeleted)))
-      .then((rows) => rows[0])
-    if (!inviterWorkspace) throw new Error(`Inviter workspace not found: ${backfill.inviterWorkspaceID}`)
-
-    const inviteeWorkspace = await tx
-      .select({ id: WorkspaceTable.id })
-      .from(WorkspaceTable)
-      .where(and(eq(WorkspaceTable.id, backfill.inviteeWorkspaceID), isNull(WorkspaceTable.timeDeleted)))
-      .then((rows) => rows[0])
-    if (!inviteeWorkspace) throw new Error(`Invitee workspace not found: ${backfill.inviteeWorkspaceID}`)
-
-    const inviteeUser = await tx
-      .select({ id: UserTable.id })
-      .from(UserTable)
-      .where(
-        and(
-          eq(UserTable.workspaceID, backfill.inviteeWorkspaceID),
-          eq(UserTable.accountID, backfill.inviteeAccountID),
-          eq(UserTable.role, "admin"),
-          isNull(UserTable.timeDeleted),
-        ),
-      )
-      .then((rows) => rows[0])
-    if (!inviteeUser) throw new Error(`Invitee workspace owner not found: ${backfill.inviteeAccountID}`)
-
-    const inviterUser = await tx
-      .select({ id: UserTable.id })
-      .from(UserTable)
-      .where(
-        and(
-          eq(UserTable.workspaceID, backfill.inviterWorkspaceID),
-          eq(UserTable.accountID, backfill.inviteeAccountID),
-          isNull(UserTable.timeDeleted),
-        ),
-      )
-      .then((rows) => rows[0])
-    if (inviterUser) throw new Error(`Self-referral is not allowed: ${backfill.inviteeAccountID}`)
-
-    const lite = await tx
-      .select({ id: LiteTable.id })
-      .from(LiteTable)
-      .where(
-        and(
-          eq(LiteTable.workspaceID, backfill.inviteeWorkspaceID),
-          eq(LiteTable.userID, inviteeUser.id),
-          isNull(LiteTable.timeDeleted),
-        ),
-      )
-      .then((rows) => rows[0])
-    if (!lite) throw new Error(`Invitee Lite subscription not found: ${backfill.inviteeWorkspaceID}`)
-
-    const existingReferral = await tx
-      .select({ id: ReferralTable.id, workspaceID: ReferralTable.workspaceID })
-      .from(ReferralTable)
-      .where(and(eq(ReferralTable.inviteeAccountID, backfill.inviteeAccountID), isNull(ReferralTable.timeDeleted)))
-      .then((rows) => rows[0])
-    if (existingReferral && existingReferral.workspaceID !== backfill.inviterWorkspaceID) {
-      throw new Error(`Referral already belongs to ${existingReferral.workspaceID}: ${existingReferral.id}`)
-    }
-
-    const referralID = existingReferral?.id ?? Identifier.create("referral")
-    if (!existingReferral) {
-      await tx.insert(ReferralTable).ignore().values({
-        workspaceID: backfill.inviterWorkspaceID,
-        id: referralID,
-        inviteeAccountID: backfill.inviteeAccountID,
-      })
-
-      const referral = await tx
-        .select({ id: ReferralTable.id })
-        .from(ReferralTable)
-        .where(and(eq(ReferralTable.inviteeAccountID, backfill.inviteeAccountID), isNull(ReferralTable.timeDeleted)))
-        .then((rows) => rows[0])
-      if (!referral) throw new Error(`Referral not created: ${backfill.inviteeAccountID}`)
-      if (referral.id !== referralID) throw new Error(`Referral already redeemed: ${referral.id}`)
-    }
-
-    const rewardInsert = await tx
-      .insert(ReferralRewardTable)
-      .ignore()
-      .values([
-        {
-          workspaceID: backfill.inviterWorkspaceID,
-          referralID,
-          amount: Referral.REWARD_AMOUNT,
-        },
-        {
-          workspaceID: backfill.inviteeWorkspaceID,
-          referralID,
-          amount: Referral.REWARD_AMOUNT,
-        },
-      ])
-
-    const rewards = await tx
-      .select({ workspaceID: ReferralRewardTable.workspaceID, amount: ReferralRewardTable.amount })
-      .from(ReferralRewardTable)
-      .where(
-        and(
-          eq(ReferralRewardTable.referralID, referralID),
-          inArray(ReferralRewardTable.workspaceID, [backfill.inviterWorkspaceID, backfill.inviteeWorkspaceID]),
-          isNull(ReferralRewardTable.timeDeleted),
-        ),
-      )
-    if (rewards.length !== 2) throw new Error(`Referral rewards not created: ${referralID}`)
-    if (rewards.some((reward) => reward.amount !== Referral.REWARD_AMOUNT)) {
-      throw new Error(`Referral reward amount mismatch: ${referralID}`)
-    }
-
-    return {
-      referralID,
-      createdReferral: !existingReferral,
-      createdRewards: rewardInsert.rowsAffected,
-      inviteeUserID: inviteeUser.id,
-      liteID: lite.id,
-      rewardWorkspaces: rewards.map((reward) => reward.workspaceID),
-    }
-  })
-
-  console.log(`  invitee user: ${result.inviteeUserID}`)
-  console.log(`  lite: ${result.liteID}`)
-  console.log(`  referral: ${result.referralID} (${result.createdReferral ? "created" : "existing"})`)
-  console.log(`  rewards: ${result.rewardWorkspaces.join(", ")} (${result.createdRewards} inserted)`)
-}
-
-console.log("Referral backfill complete")

packages/console/core/src/referral.ts

@@ -325,13 +325,6 @@ export namespace Referral {
         .then((rows) => rows.map((row) => row.workspaceID))
       if (workspaceIDs.length === 0) return
 
-      const activeLite = await tx
-        .select({ id: LiteTable.id })
-        .from(LiteTable)
-        .where(and(inArray(LiteTable.workspaceID, workspaceIDs), isNull(LiteTable.timeDeleted)))
-        .then((rows) => rows[0])
-      if (activeLite) return
-
       const litePayment = await tx
         .select({ id: PaymentTable.id })
         .from(PaymentTable)
@@ -384,30 +377,23 @@ export namespace Referral {
         .then((rows) => rows[0])
       if (!referral) return
 
-      await tx.insert(ReferralRewardTable).ignore().values({
-        workspaceID: referral.workspaceID,
-        referralID: referral.id,
-        amount: REWARD_AMOUNT,
-      })
+      const result = await tx
+        .insert(ReferralRewardTable)
+        .ignore()
+        .values([
+          {
+            workspaceID: referral.workspaceID,
+            referralID: referral.id,
+            amount: REWARD_AMOUNT,
+          },
+          {
+            workspaceID: input.workspaceID,
+            referralID: referral.id,
+            amount: REWARD_AMOUNT,
+          },
+        ])
 
-      const existingInviteeReward = await tx
-        .select({ workspaceID: ReferralRewardTable.workspaceID })
-        .from(ReferralRewardTable)
-        .where(
-          and(
-            eq(ReferralRewardTable.referralID, referral.id),
-            sql`${ReferralRewardTable.workspaceID} <> ${referral.workspaceID}`,
-            isNull(ReferralRewardTable.timeDeleted),
-          ),
-        )
-        .then((rows) => rows[0])
-      if (existingInviteeReward) return
-
-      await tx.insert(ReferralRewardTable).ignore().values({
-        workspaceID: input.workspaceID,
-        referralID: referral.id,
-        amount: REWARD_AMOUNT,
-      })
+      if (result.rowsAffected === 0) return
     })
   }
 

packages/console/function/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@opencode-ai/console-function",
-  "version": "1.15.13",
+  "version": "1.15.11",
   "$schema": "https://json.schemastore.org/package.json",
   "private": true,
   "type": "module",

packages/console/function/src/log-processor.ts

@@ -21,7 +21,6 @@ export default {
       )
         continue
 
-      const ip = event.event.request.headers["x-real-ip"]
       let data: Record<string, unknown> = {
         "cf.continent": event.event.request.cf?.continent,
         "cf.country": event.event.request.cf?.country,
@@ -33,8 +32,7 @@ export default {
         duration: event.wallTime,
         request_length: parseInt(event.event.request.headers["content-length"] ?? "0"),
         status: event.event.response?.status ?? 0,
-        ip,
-        "ip.prefix": ipPrefix(ip),
+        ip: event.event.request.headers["x-real-ip"],
       }
       const time = new Date(event.eventTimestamp ?? Date.now()).toISOString()
       const events = [
@@ -112,7 +110,6 @@ function toLakeEvent(time: string, data: Record<string, unknown>) {
     request_length: integer(data, "request_length"),
     status: integer(data, "status"),
     ip: string(data, "ip"),
-    ip_prefix: string(data, "ip.prefix"),
     is_stream: boolean(data, "is_stream"),
     session: string(data, "session"),
     request: string(data, "request"),
@@ -149,35 +146,15 @@ function toLakeEvent(time: string, data: Record<string, unknown>) {
     cost_cache_read_microcents: integer(data, "cost.cache_read.microcents"),
     cost_cache_write_microcents: integer(data, "cost.cache_write.microcents"),
     cost_total_microcents: integer(data, "cost.total.microcents"),
+    cost_input: integer(data, "cost.input"),
+    cost_output: integer(data, "cost.output"),
+    cost_cache_read: integer(data, "cost.cache_read"),
+    cost_cache_write_5m: integer(data, "cost.cache_write_5m"),
+    cost_cache_write_1h: integer(data, "cost.cache_write_1h"),
+    cost_total: integer(data, "cost.total"),
   }
 }
 
-// Returns a stable lookup key for an IP address.
-// IPv4: full address as /32 (e.g. "203.0.113.45/32").
-// IPv6: the /64 network prefix (e.g. "2001:db8:abcd:1234::/64"). ISPs commonly
-// rotate the lower 64 host bits via SLAAC privacy extensions (RFC 8981), so
-// grouping by /64 collapses those rotations into one key.
-function ipPrefix(ip: string | undefined) {
-  if (!ip) return undefined
-  if (ip.includes(".") && !ip.includes(":")) return `${ip}/32`
-  if (!ip.includes(":")) return undefined
-
-  // Expand "::" to its full form, then keep the first 4 hextets.
-  const [head, tail] = ip.split("::") as [string, string | undefined]
-  const headParts = head ? head.split(":") : []
-  const tailParts = tail !== undefined ? tail.split(":") : []
-  const missing = 8 - headParts.length - tailParts.length
-  if (missing < 0) return undefined
-  const full = [...headParts, ...new Array(missing).fill("0"), ...tailParts]
-  if (full.length !== 8) return undefined
-
-  const prefix = full
-    .slice(0, 4)
-    .map((part) => part.toLowerCase().replace(/^0+(?=.)/, ""))
-    .join(":")
-  return `${prefix}::/64`
-}
-
 function string(data: Record<string, unknown>, key: string) {
   const value = data[key]
   if (typeof value === "string") return value

packages/console/mail/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@opencode-ai/console-mail",
-  "version": "1.15.13",
+  "version": "1.15.11",
   "dependencies": {
     "@jsx-email/all": "2.2.3",
     "@jsx-email/cli": "1.4.3",

packages/console/support/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@opencode-ai/console-support",
-  "version": "1.15.13",
+  "version": "1.15.11",
   "type": "module",
   "license": "MIT",
   "scripts": {

packages/core/package.json

@@ -1,6 +1,6 @@
 {
   "$schema": "https://json.schemastore.org/package.json",
-  "version": "1.15.13",
+  "version": "1.15.11",
   "name": "@opencode-ai/core",
   "type": "module",
   "license": "MIT",
@@ -59,11 +59,10 @@
     "ai-gateway-provider": "3.1.2",
     "cross-spawn": "catalog:",
     "effect": "catalog:",
-    "gitlab-ai-provider": "6.8.0",
+    "gitlab-ai-provider": "6.7.0",
     "glob": "13.0.5",
     "google-auth-library": "10.5.0",
     "immer": "11.1.4",
-    "jsonc-parser": "3.3.1",
     "mime-types": "3.0.2",
     "minimatch": "10.2.5",
     "npm-package-arg": "13.0.2",

packages/core/src/agent.ts

@@ -1,107 +1,147 @@
 export * as AgentV2 from "./agent"
 
-import { Array, Context, Effect, Layer, Schema, Scope } from "effect"
-import { castDraft, enableMapSet, type Draft } from "immer"
+import { Context, Effect, HashMap, Layer, Option, Order, pipe, Schema, Array } from "effect"
+import { produce, type Draft } from "immer"
 import { ModelV2 } from "./model"
 import { PermissionV2 } from "./permission"
+import { PluginV2 } from "./plugin"
 import { ProviderV2 } from "./provider"
-import { PositiveInt } from "./schema"
-import { State } from "./state"
 
 export const ID = Schema.String.pipe(Schema.brand("AgentV2.ID"))
 export type ID = typeof ID.Type
 
-export const Color = Schema.Union([
-  Schema.String.check(Schema.isPattern(/^#[0-9a-fA-F]{6}$/)),
-  Schema.Literals(["primary", "secondary", "accent", "success", "warning", "error", "info"]),
-])
+export const Mode = Schema.Literals(["subagent", "primary", "all"]).annotate({ identifier: "AgentV2.Mode" })
+export type Mode = typeof Mode.Type
 
-export class Info extends Schema.Class<Info>("AgentV2.Info")({
-  id: ID,
+export const Info = Schema.Struct({
+  name: ID,
+  description: Schema.optional(Schema.String),
+  mode: Mode,
+  hidden: Schema.Boolean.pipe(Schema.optional),
+  color: Schema.String.pipe(Schema.optional),
+  permission: PermissionV2.Ruleset,
   model: ModelV2.Ref.pipe(Schema.optional),
-  options: ProviderV2.Options,
   system: Schema.String.pipe(Schema.optional),
-  description: Schema.String.pipe(Schema.optional),
-  mode: Schema.Literals(["subagent", "primary", "all"]),
-  hidden: Schema.Boolean,
-  color: Color.pipe(Schema.optional),
-  steps: PositiveInt.pipe(Schema.optional),
-  permissions: PermissionV2.Ruleset,
-}) {
-  static empty(id: ID) {
-    return new Info({
-      id,
-      options: {
-        headers: {},
-        body: {},
-        aisdk: {
-          provider: {},
-          request: {},
-        },
-      },
-      mode: "all",
-      hidden: false,
-      permissions: [],
-    })
-  }
-}
+  options: ProviderV2.Options.pipe(Schema.optional),
+  steps: Schema.Int.pipe(Schema.optional),
+}).annotate({ identifier: "AgentV2.Info" })
+export type Info = typeof Info.Type
 
-type Data = {
-  agents: Map<ID, Info>
-}
+export class NotFoundError extends Schema.TaggedErrorClass<NotFoundError>()("AgentV2.NotFound", {
+  agent: ID,
+}) {}
 
-export type Editor = {
-  list: () => readonly Info[]
-  get: (id: ID) => Info | undefined
-  update: (id: ID, fn: (agent: Draft<Info>) => void) => void
-  remove: (id: ID) => void
-}
+export class InvalidDefaultError extends Schema.TaggedErrorClass<InvalidDefaultError>()("AgentV2.InvalidDefault", {
+  agent: ID,
+  reason: Schema.Literals(["missing", "subagent", "hidden"]),
+}) {}
+
+export class NoDefaultError extends Schema.TaggedErrorClass<NoDefaultError>()("AgentV2.NoDefault", {}) {}
 
 export interface Interface {
-  readonly transform: State.Interface<Data, Editor>["transform"]
-  readonly update: (update: State.Transform<Editor>) => Effect.Effect<void, never, Scope.Scope>
-  readonly get: (id: ID) => Effect.Effect<Info | undefined>
-  readonly all: () => Effect.Effect<Info[]>
+  readonly get: (agent: ID) => Effect.Effect<Info, NotFoundError>
+  readonly list: () => Effect.Effect<Info[]>
+  readonly update: (agent: ID, fn: (agent: Draft<Info>) => void) => Effect.Effect<void>
+  readonly remove: (agent: ID) => Effect.Effect<void>
+  readonly defaultInfo: () => Effect.Effect<Info, InvalidDefaultError | NoDefaultError>
+  readonly defaultAgent: () => Effect.Effect<ID, InvalidDefaultError | NoDefaultError>
+  readonly setDefault: (agent: ID) => Effect.Effect<void, NotFoundError>
 }
 
 export class Service extends Context.Service<Service, Interface>()("@opencode/v2/Agent") {}
 
-enableMapSet()
-
 export const layer = Layer.effect(
   Service,
   Effect.gen(function* () {
-    const state = State.create<Data, Editor>({
-      initial: () => ({ agents: new Map() }),
-      editor: (draft) => ({
-        list: () => Array.fromIterable(draft.agents.values()) as Info[],
-        get: (id) => draft.agents.get(id),
-        update: (id, fn) => {
-          const current = draft.agents.get(id) ?? castDraft(Info.empty(id))
-          if (!draft.agents.has(id)) draft.agents.set(id, current)
-          fn(current)
-          current.id = id
-        },
-        remove: (id) => {
-          draft.agents.delete(id)
-        },
-      }),
-    })
+    const plugin = yield* PluginV2.Service
+    let agents = HashMap.empty<ID, Info>()
+    let defaultAgent: ID | undefined
 
-    return Service.of({
-      transform: state.transform,
-      update: Effect.fn("AgentV2.update")(function* (update) {
-        const transform = yield* state.transform()
-        yield* transform(update)
+    const result: Interface = {
+      get: Effect.fn("AgentV2.get")(function* (agent) {
+        const match = HashMap.get(agents, agent)
+        if (!match.valueOrUndefined) return yield* new NotFoundError({ agent })
+        return match.value
       }),
-      get: Effect.fn("AgentV2.get")(function* (id) {
-        return state.get().agents.get(id)
+
+      list: Effect.fn("AgentV2.list")(function* () {
+        return pipe(
+          HashMap.toValues(agents),
+          Array.sortWith((agent) => agent.name, Order.String),
+        )
       }),
-      all: Effect.fn("AgentV2.all")(function* () {
-        return Array.fromIterable(state.get().agents.values())
+
+      update: Effect.fnUntraced(function* (agent, fn) {
+        const next = produce(
+          HashMap.get(agents, agent).pipe(
+            Option.getOrElse(
+              () =>
+                ({
+                  name: agent,
+                  mode: "all",
+                  permission: [],
+                  options: {
+                    headers: {},
+                    body: {},
+                    aisdk: {
+                      provider: {},
+                      request: {},
+                    },
+                  },
+                }) satisfies Info,
+            ),
+          ),
+          fn,
+        )
+        const updated = yield* plugin.trigger("agent.update", {}, { agent: next, cancel: false })
+        if (updated.cancel) return
+        agents = HashMap.set(agents, agent, { ...updated.agent, name: agent })
       }),
-    })
+
+      remove: Effect.fn("AgentV2.remove")(function* (agent) {
+        const existing = Option.getOrUndefined(HashMap.get(agents, agent))
+        if (!existing) return
+        if ((yield* plugin.trigger("agent.remove", { agent: existing }, { cancel: false })).cancel) return
+        agents = HashMap.remove(agents, agent)
+        if (defaultAgent === agent) defaultAgent = undefined
+      }),
+
+      defaultInfo: Effect.fn("AgentV2.defaultInfo")(function* () {
+        const updated = yield* plugin.trigger("agent.default", {}, { agent: defaultAgent })
+        const selected = updated.agent
+        if (selected) {
+          const agent = yield* result
+            .get(selected)
+            .pipe(
+              Effect.catchTag("AgentV2.NotFound", () =>
+                Effect.fail(new InvalidDefaultError({ agent: selected, reason: "missing" })),
+              ),
+            )
+          if (agent.mode === "subagent") return yield* new InvalidDefaultError({ agent: selected, reason: "subagent" })
+          if (agent.hidden === true) return yield* new InvalidDefaultError({ agent: selected, reason: "hidden" })
+          return agent
+        }
+
+        const visible = pipe(
+          yield* result.list(),
+          Array.findFirst((agent) => agent.mode !== "subagent" && agent.hidden !== true),
+        )
+        if (Option.isSome(visible)) return visible.value
+        return yield* new NoDefaultError()
+      }),
+
+      defaultAgent: Effect.fn("AgentV2.defaultAgent")(function* () {
+        return (yield* result.defaultInfo()).name
+      }),
+
+      setDefault: Effect.fn("AgentV2.setDefault")(function* (agent) {
+        yield* result.get(agent)
+        defaultAgent = agent
+      }),
+    }
+
+    return Service.of(result)
   }),
 )
 
-export const defaultLayer = layer
+export const defaultLayer = layer.pipe(Layer.provide(PluginV2.defaultLayer))

packages/core/src/catalog.ts

@@ -1,22 +1,18 @@
 export * as Catalog from "./catalog"
 
-import { Context, Effect, Layer, Option, Order, pipe, Schema, Array, Scope, Stream } from "effect"
-import { castDraft, enableMapSet, type Draft } from "immer"
+import { Context, Effect, HashMap, Layer, Option, Order, pipe, Schema, Array, Scope, Stream } from "effect"
+import { produce, type Draft } from "immer"
 import { ModelV2 } from "./model"
 import { PluginV2 } from "./plugin"
 import { ProviderV2 } from "./provider"
 import { Location } from "./location"
 import { EventV2 } from "./event"
-import { Policy } from "./policy"
-import { State } from "./state"
 
 export type ProviderRecord = {
   provider: ProviderV2.Info
   models: Map<ModelV2.ID, ModelV2.Info>
 }
 
-export type DefaultModel = { providerID: ProviderV2.ID; modelID: ModelV2.ID }
-
 export class ProviderNotFoundError extends Schema.TaggedErrorClass<ProviderNotFoundError>()(
   "CatalogV2.ProviderNotFound",
   {
@@ -29,8 +25,6 @@ export class ModelNotFoundError extends Schema.TaggedErrorClass<ModelNotFoundErr
   modelID: ModelV2.ID,
 }) {}
 
-export const PolicyActions = Schema.Literals(["provider.use"])
-
 export const Event = {
   ModelUpdated: EventV2.define({
     type: "catalog.model.updated",
@@ -40,31 +34,24 @@ export const Event = {
   }),
 }
 
-type Data = {
-  providers: Map<ProviderV2.ID, ProviderRecord>
-  defaultModel?: DefaultModel
-}
-
-export type Editor = {
+export type Context = {
+  data: readonly ProviderRecord[]
+  updateProvider: (providerID: ProviderV2.ID, fn: (provider: Draft<ProviderV2.Info>) => void) => void
+  updateModel: (providerID: ProviderV2.ID, modelID: ModelV2.ID, fn: (model: Draft<ModelV2.Info>) => void) => void
   provider: {
-    list: () => readonly ProviderRecord[]
-    get: (providerID: ProviderV2.ID) => ProviderRecord | undefined
     update: (providerID: ProviderV2.ID, fn: (provider: Draft<ProviderV2.Info>) => void) => void
     remove: (providerID: ProviderV2.ID) => void
   }
   model: {
-    get: (providerID: ProviderV2.ID, modelID: ModelV2.ID) => ModelV2.Info | undefined
     update: (providerID: ProviderV2.ID, modelID: ModelV2.ID, fn: (model: Draft<ModelV2.Info>) => void) => void
     remove: (providerID: ProviderV2.ID, modelID: ModelV2.ID) => void
-    default: {
-      get: () => DefaultModel | undefined
-      set: (providerID: ProviderV2.ID, modelID: ModelV2.ID) => void
-    }
   }
 }
 
+export type Loader = (update: (ctx: Context) => void) => Effect.Effect<void>
+
 export interface Interface {
-  readonly transform: State.Interface<Data, Editor>["transform"]
+  readonly loader: () => Effect.Effect<Loader, never, Scope.Scope>
   readonly provider: {
     readonly get: (providerID: ProviderV2.ID) => Effect.Effect<ProviderV2.Info, ProviderNotFoundError>
     readonly all: () => Effect.Effect<ProviderV2.Info[]>
@@ -78,25 +65,29 @@ export interface Interface {
     readonly all: () => Effect.Effect<ModelV2.Info[]>
     readonly available: () => Effect.Effect<ModelV2.Info[]>
     readonly default: () => Effect.Effect<Option.Option<ModelV2.Info>>
+    readonly setDefault: (
+      providerID: ProviderV2.ID,
+      modelID: ModelV2.ID,
+    ) => Effect.Effect<void, ProviderNotFoundError | ModelNotFoundError>
     readonly small: (providerID: ProviderV2.ID) => Effect.Effect<Option.Option<ModelV2.Info>>
   }
 }
 
 export class Service extends Context.Service<Service, Interface>()("@opencode/v2/Catalog") {}
 
-enableMapSet()
-
 export const layer = Layer.effect(
   Service,
   Effect.gen(function* () {
     yield* Location.Service
+    let records = HashMap.empty<ProviderV2.ID, ProviderRecord>()
+    let loaders: { update: (ctx: Context) => void }[] = []
+    let defaultModel: { providerID: ProviderV2.ID; modelID: ModelV2.ID } | undefined
     const plugin = yield* PluginV2.Service
     const events = yield* EventV2.Service
-    const policy = yield* Policy.Service
     const scope = yield* Scope.Scope
 
     const resolve = (model: ModelV2.Info) => {
-      const provider = state.get().providers.get(model.providerID)!.provider
+      const provider = Option.getOrThrow(HashMap.get(records, model.providerID)).provider
       const endpoint =
         model.endpoint.type === "unknown"
           ? provider.endpoint
@@ -129,9 +120,9 @@ export const layer = Layer.effect(
     }
 
     function* getRecord(providerID: ProviderV2.ID) {
-      const match = state.get().providers.get(providerID)
-      if (!match) return yield* new ProviderNotFoundError({ providerID })
-      return match
+      const match = HashMap.get(records, providerID)
+      if (!match.valueOrUndefined) return yield* new ProviderNotFoundError({ providerID })
+      return match.value
     }
 
     const normalizeEndpoint = (item: Draft<ProviderV2.Info> | Draft<ModelV2.Info>) => {
@@ -140,73 +131,114 @@ export const layer = Layer.effect(
       delete item.options.aisdk.provider.baseURL
     }
 
-    const state = State.create<Data, Editor>({
-      initial: () => ({ providers: new Map() }),
-      editor: (draft) => {
-        const result: Editor = {
-          provider: {
-            list: () => Array.fromIterable(draft.providers.values()) as ProviderRecord[],
-            get: (providerID) => draft.providers.get(providerID),
-            update: (providerID, fn) => {
-              let current = draft.providers.get(providerID)
-              if (!current) {
-                current = castDraft({
-                  provider: ProviderV2.Info.empty(providerID),
-                  models: new Map<ModelV2.ID, ModelV2.Info>(),
-                })
-                draft.providers.set(providerID, current)
-              }
-              fn(current.provider)
-              normalizeEndpoint(current.provider)
-            },
-            remove: (providerID) => {
-              draft.providers.delete(providerID)
-            },
+    const clone = (input: HashMap.HashMap<ProviderV2.ID, ProviderRecord>) =>
+      HashMap.fromIterable(
+        HashMap.toEntries(input).map(([key, value]) => [key, { ...value, models: new Map(value.models) }] as const),
+      )
+
+    const context = (draft: {
+      records: HashMap.HashMap<ProviderV2.ID, ProviderRecord>
+      data: ProviderRecord[]
+    }): Context => {
+      const result: Context = {
+        data: draft.data,
+        updateProvider: (providerID, fn) => result.provider.update(providerID, fn),
+        updateModel: (providerID, modelID, fn) => result.model.update(providerID, modelID, fn),
+        provider: {
+          update: (providerID, fn) => {
+            const current = Option.getOrUndefined(HashMap.get(draft.records, providerID))
+            const provider = produce(current?.provider ?? ProviderV2.Info.empty(providerID), (draft) => {
+              fn(draft)
+              normalizeEndpoint(draft)
+            })
+            const next = {
+              provider,
+              models: current?.models ?? new Map<ModelV2.ID, ModelV2.Info>(),
+            }
+            draft.records = HashMap.set(draft.records, providerID, next)
+            const index = draft.data.findIndex((item) => item.provider.id === providerID)
+            if (index === -1) draft.data.push(next)
+            else draft.data[index] = next
           },
-          model: {
-            get: (providerID, modelID) => draft.providers.get(providerID)?.models.get(modelID),
-            update: (providerID, modelID, fn) => {
-              result.provider.update(providerID, () => {})
-              const record = draft.providers.get(providerID)!
-              const model = record.models.get(modelID) ?? castDraft(ModelV2.Info.empty(providerID, modelID))
-              if (!record.models.has(modelID)) record.models.set(modelID, model)
-              fn(model)
-              model.id = modelID
-              model.providerID = providerID
-              normalizeEndpoint(model)
-            },
-            remove: (providerID, modelID) => {
-              draft.providers.get(providerID)?.models.delete(modelID)
-            },
-            default: {
-              get: () => draft.defaultModel,
-              set: (providerID, modelID) => {
-                draft.defaultModel = { providerID, modelID }
-              },
-            },
+          remove: (providerID) => {
+            draft.records = HashMap.remove(draft.records, providerID)
+            const index = draft.data.findIndex((item) => item.provider.id === providerID)
+            if (index !== -1) draft.data.splice(index, 1)
           },
-        }
-        return result
-      },
-      finalize: Effect.fn("CatalogV2.finalize")(function* (catalog, reason) {
-        if (reason !== "plugin.added") yield* plugin.trigger("catalog.transform", catalog, {}).pipe(Effect.asVoid)
-        for (const record of [...catalog.provider.list()]) {
-          if ((yield* policy.evaluate("provider.use", record.provider.id, "allow")) === "deny") {
-            catalog.provider.remove(record.provider.id)
-          }
-        }
-      }),
+        },
+        model: {
+          update: (providerID, modelID, fn) => {
+            const current = Option.getOrThrow(HashMap.get(draft.records, providerID))
+            const model = produce(current.models.get(modelID) ?? ModelV2.Info.empty(providerID, modelID), (draft) => {
+              fn(draft)
+              normalizeEndpoint(draft)
+            })
+            const next = {
+              provider: current.provider,
+              models: new Map(current.models).set(modelID, new ModelV2.Info({ ...model, id: modelID, providerID })),
+            }
+            draft.records = HashMap.set(draft.records, providerID, next)
+            const index = draft.data.findIndex((item) => item.provider.id === providerID)
+            if (index === -1) draft.data.push(next)
+            else draft.data[index] = next
+          },
+          remove: (providerID, modelID) => {
+            const current = Option.getOrUndefined(HashMap.get(draft.records, providerID))
+            if (!current) return
+            const next = {
+              provider: current.provider,
+              models: new Map(current.models),
+            }
+            next.models.delete(modelID)
+            draft.records = HashMap.set(draft.records, providerID, next)
+            const index = draft.data.findIndex((item) => item.provider.id === providerID)
+            if (index !== -1) draft.data[index] = next
+          },
+        },
+      }
+      return result
+    }
+
+    const transform = Effect.fn("CatalogV2.transform")(function* () {
+      const draft = { records: clone(records), data: HashMap.toValues(records) }
+      yield* plugin.trigger("catalog.transform", context(draft), {})
+      records = draft.records
     })
 
-    yield* events.subscribe(PluginV2.Event.Added).pipe(
-      Stream.runForEach((event) =>
-        state.update((catalog) => plugin.triggerFor(event.data.id, "catalog.transform", catalog, {}), "plugin.added"),
+    const rebuild = Effect.fn("CatalogV2.rebuild")(function* () {
+      const draft = { records: HashMap.empty<ProviderV2.ID, ProviderRecord>(), data: [] as ProviderRecord[] }
+      for (const loader of loaders) loader.update(context(draft))
+      yield* plugin.trigger("catalog.transform", context(draft), {})
+      records = draft.records
+    })
+
+    yield* plugin.added().pipe(
+      Stream.runForEach((id) =>
+        Effect.gen(function* () {
+          const draft = { records: clone(records), data: HashMap.toValues(records) }
+          yield* plugin.triggerFor(id, "catalog.transform", context(draft), {})
+          records = draft.records
+        }),
       ),
       Effect.forkIn(scope, { startImmediately: true }),
     )
 
     const result: Interface = {
-      transform: state.transform,
+      loader: Effect.fn("CatalogV2.loader")(function* () {
+        const loader = { update: (_ctx: Context) => {} }
+        loaders = [...loaders, loader]
+        const scope = yield* Scope.Scope
+        yield* Scope.addFinalizer(
+          scope,
+          Effect.sync(() => {
+            loaders = loaders.filter((item) => item !== loader)
+          }).pipe(Effect.andThen(rebuild())),
+        )
+        return Effect.fnUntraced(function* (update) {
+          loader.update = update
+          yield* rebuild()
+        })
+      }),
 
       provider: {
         get: Effect.fn("CatalogV2.provider.get")(function* (providerID) {
@@ -215,11 +247,11 @@ export const layer = Layer.effect(
         }),
 
         all: Effect.fn("CatalogV2.provider.all")(function* () {
-          return Array.fromIterable(state.get().providers.values()).map((record) => record.provider)
+          return globalThis.Array.from(HashMap.values(records)).map((record) => record.provider)
         }),
 
         available: Effect.fn("CatalogV2.provider.available")(function* () {
-          return Array.fromIterable(state.get().providers.values())
+          return globalThis.Array.from(HashMap.values(records))
             .map((record) => record.provider)
             .filter((provider) => provider.enabled)
         }),
@@ -235,8 +267,9 @@ export const layer = Layer.effect(
 
         all: Effect.fn("CatalogV2.model.all")(function* () {
           return pipe(
-            Array.fromIterable(state.get().providers.values()),
-            Array.flatMap((record) => Array.fromIterable(record.models.values())),
+            records,
+            HashMap.toValues,
+            Array.flatMap((record) => globalThis.Array.from(record.models.values())),
             Array.map(resolve),
             Array.sortWith((item) => item.time.released.epochMilliseconds, Order.flip(Order.Number)),
           )
@@ -244,13 +277,12 @@ export const layer = Layer.effect(
 
         available: Effect.fn("CatalogV2.model.available")(function* () {
           return (yield* result.model.all()).filter((model) => {
-            const record = state.get().providers.get(model.providerID)
+            const record = Option.getOrUndefined(HashMap.get(records, model.providerID))
             return record?.provider.enabled !== false && model.enabled
           })
         }),
 
         default: Effect.fn("CatalogV2.model.default")(function* () {
-          const defaultModel = state.get().defaultModel
           if (defaultModel) {
             const model = yield* result.model.get(defaultModel.providerID, defaultModel.modelID).pipe(Effect.option)
             if (Option.isSome(model) && model.value.enabled) return model
@@ -263,8 +295,13 @@ export const layer = Layer.effect(
           )
         }),
 
+        setDefault: Effect.fn("CatalogV2.model.setDefault")(function* (providerID, modelID) {
+          yield* result.model.get(providerID, modelID)
+          defaultModel = { providerID, modelID }
+        }),
+
         small: Effect.fn("CatalogV2.model.small")(function* (providerID) {
-          const record = state.get().providers.get(providerID)
+          const record = Option.getOrUndefined(HashMap.get(records, providerID))
           if (!record) return Option.none<ModelV2.Info>()
 
           if (providerID === ProviderV2.ID.opencode) {
@@ -273,7 +310,7 @@ export const layer = Layer.effect(
           }
 
           const candidates = pipe(
-            Array.fromIterable(record.models.values()),
+            globalThis.Array.from(record.models.values()),
             Array.filter(
               (model) =>
                 model.providerID === providerID &&

packages/core/src/config.ts

@@ -1,203 +0,0 @@
-export * as Config from "./config"
-
-import path from "path"
-import { type ParseError, parse } from "jsonc-parser"
-import { Context, Effect, Layer, Option, Schema } from "effect"
-import { AppFileSystem } from "./filesystem"
-import { Global } from "./global"
-import { Location } from "./location"
-import { PermissionV2 } from "./permission"
-import { Policy } from "./policy"
-import { AbsolutePath } from "./schema"
-import { ConfigAgent } from "./config/agent"
-import { ConfigAttachments } from "./config/attachments"
-import { ConfigCompaction } from "./config/compaction"
-import { ConfigExperimental } from "./config/experimental"
-import { ConfigFormatter } from "./config/formatter"
-import { ConfigLSP } from "./config/lsp"
-import { ConfigMCP } from "./config/mcp"
-import { ConfigPlugin } from "./config/plugin"
-import { ConfigProvider } from "./config/provider"
-import { ConfigReference } from "./config/reference"
-import { ConfigToolOutput } from "./config/tool-output"
-import { ConfigWatcher } from "./config/watcher"
-
-export class Info extends Schema.Class<Info>("Config.Info")({
-  $schema: Schema.optional(Schema.String).annotate({
-    description: "JSON schema reference for configuration validation",
-  }),
-  shell: Schema.String.pipe(Schema.optional).annotate({
-    description: "Default shell to use for terminal and shell tool execution",
-  }),
-  model: Schema.String.pipe(Schema.optional).annotate({
-    description: "Default model to use when no session or agent model is selected",
-  }),
-  autoupdate: Schema.Union([Schema.Boolean, Schema.Literal("notify")])
-    .pipe(Schema.optional)
-    .annotate({
-      description: "Automatically update or notify when a new version is available",
-    }),
-  share: Schema.Literals(["manual", "auto", "disabled"]).pipe(Schema.optional).annotate({
-    description: "Control whether sessions may be shared manually, automatically, or not at all",
-  }),
-  enterprise: Schema.Struct({
-    url: Schema.String.pipe(Schema.optional),
-  })
-    .pipe(Schema.optional)
-    .annotate({
-      description: "Enterprise sharing service configuration",
-    }),
-  username: Schema.String.pipe(Schema.optional).annotate({
-    description: "Username displayed in conversations and used for telemetry identity",
-  }),
-  permissions: PermissionV2.Ruleset.pipe(Schema.optional).annotate({
-    description: "Ordered tool permission rules applied to agent tool use",
-  }),
-  agents: Schema.Record(Schema.String, ConfigAgent.Info).pipe(Schema.optional).annotate({
-    description: "Named built-in agent overrides and custom agent definitions",
-  }),
-  snapshots: Schema.Boolean.pipe(Schema.optional).annotate({
-    description: "Enable snapshots used for undo and revert behavior",
-  }),
-  watcher: ConfigWatcher.Info.pipe(Schema.optional).annotate({
-    description: "Filesystem watcher configuration",
-  }),
-  formatter: ConfigFormatter.Info.pipe(Schema.optional).annotate({
-    description: "Enable built-in formatters or configure formatter overrides",
-  }),
-  lsp: ConfigLSP.Info.pipe(Schema.optional).annotate({
-    description: "Enable built-in language servers or configure server overrides",
-  }),
-  attachments: ConfigAttachments.Info.pipe(Schema.optional).annotate({
-    description: "Attachment processing configuration",
-  }),
-  tool_output: ConfigToolOutput.Info.pipe(Schema.optional).annotate({
-    description: "Tool output truncation thresholds",
-  }),
-  mcp: ConfigMCP.Info.pipe(Schema.optional).annotate({
-    description: "MCP server configuration",
-  }),
-  compaction: ConfigCompaction.Info.pipe(Schema.optional).annotate({
-    description: "Conversation compaction behavior",
-  }),
-  skills: Schema.String.pipe(Schema.Array, Schema.optional).annotate({
-    description: "Additional paths or URLs to discover skills from",
-  }),
-  instructions: Schema.String.pipe(Schema.Array, Schema.optional).annotate({
-    description: "Additional paths or URLs supplying ambient instructions",
-  }),
-  references: ConfigReference.Info.pipe(Schema.optional).annotate({
-    description: "Named local directories or Git repositories available as external context",
-  }),
-  plugins: ConfigPlugin.Plugins.pipe(Schema.optional).annotate({
-    description: "Ordered external plugin packages to load",
-  }),
-  experimental: ConfigExperimental.Experimental.pipe(Schema.optional),
-  providers: Schema.Record(Schema.String, ConfigProvider.Info).pipe(Schema.optional),
-}) {}
-
-export const FileSource = Schema.Struct({
-  type: Schema.Literal("file"),
-  path: Schema.String,
-}).annotate({ identifier: "Config.FileSource" })
-export type FileSource = typeof FileSource.Type
-
-export const MemorySource = Schema.Struct({
-  type: Schema.Literal("memory"),
-}).annotate({ identifier: "Config.MemorySource" })
-export type MemorySource = typeof MemorySource.Type
-
-export const Source = Schema.Union([FileSource, MemorySource]).pipe(Schema.toTaggedUnion("type"))
-export type Source = typeof Source.Type
-
-export class Loaded extends Schema.Class<Loaded>("Config.Loaded")({
-  source: Source,
-  info: Info,
-}) {}
-
-export interface Interface {
-  /** Returns supplemental config directories from lowest to highest priority. */
-  readonly directories: () => Effect.Effect<AbsolutePath[]>
-  /** Loads location config files from lowest to highest priority. */
-  readonly get: () => Effect.Effect<Loaded[]>
-}
-
-export class Service extends Context.Service<Service, Interface>()("@opencode/v2/Config") {}
-
-export const layer = Layer.effect(
-  Service,
-  Effect.gen(function* () {
-    const fs = yield* AppFileSystem.Service
-    const global = yield* Global.Service
-    const location = yield* Location.Service
-    const policy = yield* Policy.Service
-    const names = ["config.json", "opencode.json", "opencode.jsonc"]
-
-    const loadFile = Effect.fnUntraced(function* (filepath: string) {
-      const text = yield* fs.readFileStringSafe(filepath)
-      if (!text) return
-
-      const errors: ParseError[] = []
-      const input: unknown = parse(text, errors, { allowTrailingComma: true })
-      if (errors.length) return
-
-      // Accept legacy fields while v2 is migrated incrementally; recognized
-      // fields still have to satisfy the v2 schema.
-      const info = Option.getOrUndefined(
-        Schema.decodeUnknownOption(Info)(input, { errors: "all", onExcessProperty: "ignore" }),
-      )
-      if (!info) return
-      return new Loaded({ source: { type: "file", path: filepath }, info })
-    })
-
-    const loadDirectory = Effect.fnUntraced(function* (directory: AbsolutePath) {
-      return yield* Effect.forEach(names, (file) => loadFile(path.join(directory, file))).pipe(
-        Effect.map((configs) => configs.filter((config): config is Loaded => config !== undefined)),
-      )
-    })
-
-    const globalDirectory = AbsolutePath.make(global.config)
-    const locationIsGlobal = path.resolve(location.directory) === path.resolve(global.config)
-    // Read configuration once when this location opens. Later calls reuse these
-    // values until the location is reopened.
-    const directories = locationIsGlobal
-      ? [globalDirectory]
-      : [
-          globalDirectory,
-          ...(yield* fs
-            .up({ targets: [".opencode"], start: location.directory, stop: location.project.directory })
-            .pipe(Effect.orDie))
-            .toReversed()
-            .map((directory) => AbsolutePath.make(directory)),
-        ]
-    // A config closer to the opened directory should win over one higher up.
-    // Search starts nearby, so reverse the results before applying them.
-    const directPaths = locationIsGlobal
-      ? []
-      : (yield* fs
-          .up({ targets: names.toReversed(), start: location.directory, stop: location.project.directory })
-          .pipe(Effect.orDie)).toReversed()
-    const direct = yield* Effect.forEach(directPaths, loadFile).pipe(
-      Effect.orDie,
-      Effect.map((configs) => configs.filter((config): config is Loaded => config !== undefined)),
-    )
-    const supplementary = yield* Effect.forEach(directories, loadDirectory).pipe(Effect.orDie)
-    // Apply general settings first and more specific settings last:
-    // global config, project files, then `.opencode` files.
-    const configs = [...(supplementary[0] ?? []), ...direct, ...supplementary.slice(1).flat()]
-    // Rules use the opposite order so a user-global rule can override a
-    // repository rule. Statement order inside each file stays unchanged.
-    yield* policy.load(configs.toReversed().flatMap((config) => config.info.experimental?.policies ?? []))
-
-    return Service.of({
-      directories: Effect.fn("Config.directories")(function* () {
-        return directories
-      }),
-      get: Effect.fn("Config.get")(function* () {
-        return configs
-      }),
-    })
-  }),
-)
-
-export const defaultLayer = layer.pipe(Layer.provide(AppFileSystem.defaultLayer), Layer.provide(Global.defaultLayer))

packages/core/src/config/agent.ts

@@ -1,25 +0,0 @@
-export * as ConfigAgent from "./agent"
-
-import { Schema } from "effect"
-import { PermissionV2 } from "../permission"
-import { ConfigProvider } from "./provider"
-import { PositiveInt } from "../schema"
-
-export const Color = Schema.Union([
-  Schema.String.check(Schema.isPattern(/^#[0-9a-fA-F]{6}$/)),
-  Schema.Literals(["primary", "secondary", "accent", "success", "warning", "error", "info"]),
-])
-
-export class Info extends Schema.Class<Info>("ConfigV2.Agent")({
-  model: Schema.String.pipe(Schema.optional),
-  variant: Schema.String.pipe(Schema.optional),
-  options: ConfigProvider.Options.pipe(Schema.optional),
-  system: Schema.String.pipe(Schema.optional),
-  description: Schema.String.pipe(Schema.optional),
-  mode: Schema.Literals(["subagent", "primary", "all"]).pipe(Schema.optional),
-  hidden: Schema.Boolean.pipe(Schema.optional),
-  color: Color.pipe(Schema.optional),
-  steps: PositiveInt.pipe(Schema.optional),
-  disabled: Schema.Boolean.pipe(Schema.optional),
-  permissions: PermissionV2.Ruleset.pipe(Schema.optional),
-}) {}

packages/core/src/config/attachments.ts

@@ -1,15 +0,0 @@
-export * as ConfigAttachments from "./attachments"
-
-import { Schema } from "effect"
-import { PositiveInt } from "../schema"
-
-export class Image extends Schema.Class<Image>("ConfigV2.Attachments.Image")({
-  auto_resize: Schema.Boolean.pipe(Schema.optional),
-  max_width: PositiveInt.pipe(Schema.optional),
-  max_height: PositiveInt.pipe(Schema.optional),
-  max_base64_bytes: PositiveInt.pipe(Schema.optional),
-}) {}
-
-export class Info extends Schema.Class<Info>("ConfigV2.Attachments")({
-  image: Image.pipe(Schema.optional),
-}) {}

packages/core/src/config/compaction.ts

@@ -1,16 +0,0 @@
-export * as ConfigCompaction from "./compaction"
-
-import { Schema } from "effect"
-import { NonNegativeInt } from "../schema"
-
-export class Keep extends Schema.Class<Keep>("ConfigV2.Compaction.Keep")({
-  turns: NonNegativeInt.pipe(Schema.optional),
-  tokens: NonNegativeInt.pipe(Schema.optional),
-}) {}
-
-export class Info extends Schema.Class<Info>("ConfigV2.Compaction")({
-  auto: Schema.Boolean.pipe(Schema.optional),
-  prune: Schema.Boolean.pipe(Schema.optional),
-  keep: Keep.pipe(Schema.optional),
-  buffer: NonNegativeInt.pipe(Schema.optional),
-}) {}

packages/core/src/config/experimental.ts

@@ -1,18 +0,0 @@
-export * as ConfigExperimental from "./experimental"
-
-import { Schema } from "effect"
-import { Catalog } from "../catalog"
-import { Policy as PolicyV2 } from "../policy"
-
-// Each core domain exports the policy actions it supports. Adding an action to
-// this union makes it valid in authored config while keeping Policy generic.
-export const PolicyAction = Schema.Union([Catalog.PolicyActions])
-
-export class Policy extends Schema.Class<Policy>("ConfigV2.Experimental.Policy")({
-  ...PolicyV2.Info.fields,
-  action: PolicyAction,
-}) {}
-
-export class Experimental extends Schema.Class<Experimental>("ConfigV2.Experimental")({
-  policies: Policy.pipe(Schema.Array, Schema.optional),
-}) {}

packages/core/src/config/formatter.ts

@@ -1,12 +0,0 @@
-export * as ConfigFormatter from "./formatter"
-
-import { Schema } from "effect"
-
-export class Entry extends Schema.Class<Entry>("ConfigV2.Formatter.Entry")({
-  disabled: Schema.Boolean.pipe(Schema.optional),
-  command: Schema.String.pipe(Schema.Array, Schema.optional),
-  environment: Schema.Record(Schema.String, Schema.String).pipe(Schema.optional),
-  extensions: Schema.String.pipe(Schema.Array, Schema.optional),
-}) {}
-
-export const Info = Schema.Union([Schema.Boolean, Schema.Record(Schema.String, Entry)])

packages/core/src/config/lsp.ts

@@ -1,18 +0,0 @@
-export * as ConfigLSP from "./lsp"
-
-import { Schema } from "effect"
-
-export const Disabled = Schema.Struct({
-  disabled: Schema.Literal(true),
-})
-
-export class Server extends Schema.Class<Server>("ConfigV2.LSP.Server")({
-  command: Schema.String.pipe(Schema.Array),
-  extensions: Schema.String.pipe(Schema.Array, Schema.optional),
-  disabled: Schema.Boolean.pipe(Schema.optional),
-  env: Schema.Record(Schema.String, Schema.String).pipe(Schema.optional),
-  initialization: Schema.Record(Schema.String, Schema.Unknown).pipe(Schema.optional),
-}) {}
-
-export const Entry = Schema.Union([Disabled, Server])
-export const Info = Schema.Union([Schema.Boolean, Schema.Record(Schema.String, Entry)])

packages/core/src/config/mcp.ts

@@ -1,36 +0,0 @@
-export * as ConfigMCP from "./mcp"
-
-import { Schema } from "effect"
-import { PositiveInt } from "../schema"
-
-export class Local extends Schema.Class<Local>("ConfigV2.MCP.Local")({
-  type: Schema.Literal("local"),
-  command: Schema.String.pipe(Schema.Array),
-  environment: Schema.Record(Schema.String, Schema.String).pipe(Schema.optional),
-  disabled: Schema.Boolean.pipe(Schema.optional),
-  timeout: PositiveInt.pipe(Schema.optional),
-}) {}
-
-export class OAuth extends Schema.Class<OAuth>("ConfigV2.MCP.OAuth")({
-  client_id: Schema.String.pipe(Schema.optional),
-  client_secret: Schema.String.pipe(Schema.optional),
-  scope: Schema.String.pipe(Schema.optional),
-  callback_port: Schema.Int.check(Schema.isBetween({ minimum: 1, maximum: 65535 })).pipe(Schema.optional),
-  redirect_uri: Schema.String.pipe(Schema.optional),
-}) {}
-
-export class Remote extends Schema.Class<Remote>("ConfigV2.MCP.Remote")({
-  type: Schema.Literal("remote"),
-  url: Schema.String,
-  headers: Schema.Record(Schema.String, Schema.String).pipe(Schema.optional),
-  oauth: Schema.Union([OAuth, Schema.Literal(false)]).pipe(Schema.optional),
-  disabled: Schema.Boolean.pipe(Schema.optional),
-  timeout: PositiveInt.pipe(Schema.optional),
-}) {}
-
-export const Server = Schema.Union([Local, Remote]).pipe(Schema.toTaggedUnion("type"))
-
-export class Info extends Schema.Class<Info>("ConfigV2.MCP")({
-  timeout: PositiveInt.pipe(Schema.optional),
-  servers: Schema.Record(Schema.String, Server).pipe(Schema.optional),
-}) {}

packages/core/src/config/plugin.ts

@@ -1,13 +0,0 @@
-export * as ConfigPlugin from "./plugin"
-
-import { Schema } from "effect"
-
-export class Entry extends Schema.Class<Entry>("ConfigV2.Plugin.Entry")({
-  package: Schema.String,
-  options: Schema.Record(Schema.String, Schema.Unknown).pipe(Schema.optional),
-}) {}
-
-export const Plugin = Schema.Union([Schema.String, Entry])
-export type Plugin = typeof Plugin.Type
-
-export const Plugins = Plugin.pipe(Schema.Array)

packages/core/src/config/plugin/agent.ts

@@ -1,65 +0,0 @@
-export * as ConfigAgentPlugin from "./agent"
-
-import { Effect } from "effect"
-import { AgentV2 } from "../../agent"
-import { Config } from "../../config"
-import { ModelV2 } from "../../model"
-import { PermissionV2 } from "../../permission"
-import { PluginV2 } from "../../plugin"
-
-export const Plugin = PluginV2.define({
-  id: PluginV2.ID.make("config-agent"),
-  effect: Effect.gen(function* () {
-    const agent = yield* AgentV2.Service
-    const config = yield* Config.Service
-    const files = yield* config.get()
-
-    yield* agent.update((editor) => {
-      const permissions = new Map<AgentV2.ID, PermissionV2.Ruleset>()
-
-      for (const file of files) {
-        for (const [id, item] of Object.entries(file.info.agents ?? {})) {
-          const agentID = AgentV2.ID.make(id)
-          if (item.disabled) {
-            editor.remove(agentID)
-            permissions.delete(agentID)
-            continue
-          }
-
-          editor.update(agentID, (agent) => {
-            if (item.model !== undefined) {
-              const model = ModelV2.parse(item.model)
-              agent.model = { id: model.modelID, providerID: model.providerID, variant: agent.model?.variant }
-            }
-            if (item.variant !== undefined && agent.model !== undefined) {
-              agent.model.variant = ModelV2.VariantID.make(item.variant)
-            }
-            if (item.options !== undefined) {
-              Object.assign(agent.options.headers, item.options.headers ?? {})
-              Object.assign(agent.options.body, item.options.body ?? {})
-              Object.assign(agent.options.aisdk.provider, item.options.aisdk?.provider ?? {})
-              Object.assign(agent.options.aisdk.request, item.options.aisdk?.request ?? {})
-            }
-            if (item.system !== undefined) agent.system = item.system
-            if (item.description !== undefined) agent.description = item.description
-            if (item.mode !== undefined) agent.mode = item.mode
-            if (item.hidden !== undefined) agent.hidden = item.hidden
-            if (item.color !== undefined) agent.color = item.color
-            if (item.steps !== undefined) agent.steps = item.steps
-          })
-
-          if (item.permissions !== undefined) {
-            permissions.set(agentID, [...(permissions.get(agentID) ?? []), ...item.permissions])
-          }
-        }
-      }
-
-      const global = files.flatMap((file) => file.info.permissions ?? [])
-      for (const current of editor.list()) {
-        editor.update(current.id, (agent) => {
-          agent.permissions.push(...global, ...(permissions.get(current.id) ?? []))
-        })
-      }
-    })
-  }),
-})

packages/core/src/config/plugin/provider.ts

@@ -1,95 +0,0 @@
-export * as ConfigProviderPlugin from "./provider"
-
-import { Effect } from "effect"
-import { Catalog } from "../../catalog"
-import { Config } from "../../config"
-import { ModelV2 } from "../../model"
-import { PluginV2 } from "../../plugin"
-import { ProviderV2 } from "../../provider"
-
-export const Plugin = PluginV2.define({
-  id: PluginV2.ID.make("config-provider"),
-  effect: Effect.gen(function* () {
-    const catalog = yield* Catalog.Service
-    const config = yield* Config.Service
-    const transform = yield* catalog.transform()
-    const files = yield* config.get()
-
-    yield* transform((catalog) => {
-      for (const file of files) {
-        for (const [id, item] of Object.entries(file.info.providers ?? {})) {
-          const providerID = ProviderV2.ID.make(id)
-          catalog.provider.update(providerID, (provider) => {
-            if (item.name !== undefined) provider.name = item.name
-            if (item.env !== undefined) provider.env = [...item.env]
-            provider.enabled = { via: "custom", data: {} }
-            if (item.endpoint !== undefined) provider.endpoint = { ...item.endpoint }
-            if (item.options !== undefined) {
-              Object.assign(provider.options.headers, item.options.headers ?? {})
-              Object.assign(provider.options.body, item.options.body ?? {})
-              Object.assign(provider.options.aisdk.provider, item.options.aisdk?.provider ?? {})
-              Object.assign(provider.options.aisdk.request, item.options.aisdk?.request ?? {})
-            }
-          })
-
-          for (const [id, config] of Object.entries(item.models ?? {})) {
-            catalog.model.update(providerID, ModelV2.ID.make(id), (model) => {
-              if (config.api_id !== undefined) model.apiID = config.api_id
-              if (config.family !== undefined) model.family = config.family
-              if (config.name !== undefined) model.name = config.name
-              if (config.endpoint !== undefined) model.endpoint = { ...config.endpoint }
-              if (config.capabilities !== undefined) {
-                model.capabilities = {
-                  tools: config.capabilities.tools,
-                  input: [...config.capabilities.input],
-                  output: [...config.capabilities.output],
-                }
-              }
-              if (config.options !== undefined) {
-                Object.assign(model.options.headers, config.options.headers ?? {})
-                Object.assign(model.options.body, config.options.body ?? {})
-                Object.assign(model.options.aisdk.provider, config.options.aisdk?.provider ?? {})
-                Object.assign(model.options.aisdk.request, config.options.aisdk?.request ?? {})
-                if (config.options.variant !== undefined) model.options.variant = config.options.variant
-              }
-              if (config.variants !== undefined) {
-                for (const variant of config.variants) {
-                  let existing = model.variants.find((item) => item.id === variant.id)
-                  if (!existing) {
-                    existing = {
-                      id: variant.id,
-                      headers: {},
-                      body: {},
-                      aisdk: {
-                        provider: {},
-                        request: {},
-                      },
-                    }
-                    model.variants.push(existing)
-                  }
-                  Object.assign(existing.headers, variant.headers ?? {})
-                  Object.assign(existing.body, variant.body ?? {})
-                  Object.assign(existing.aisdk.provider, variant.aisdk?.provider ?? {})
-                  Object.assign(existing.aisdk.request, variant.aisdk?.request ?? {})
-                }
-              }
-              if (config.cost !== undefined) {
-                model.cost = (Array.isArray(config.cost) ? config.cost : [config.cost]).map((cost) => ({
-                  tier: cost.tier && { ...cost.tier },
-                  input: cost.input,
-                  output: cost.output,
-                  cache: {
-                    read: cost.cache?.read ?? 0,
-                    write: cost.cache?.write ?? 0,
-                  },
-                }))
-              }
-              if (config.disabled !== undefined) model.enabled = !config.disabled
-              if (config.limit !== undefined) model.limit = { ...model.limit, ...config.limit }
-            })
-          }
-        }
-      }
-    })
-  }),
-})

packages/core/src/config/provider.ts

@@ -1,62 +0,0 @@
-export * as ConfigProvider from "./provider"
-
-import { Schema } from "effect"
-import { ProviderV2 } from "../provider"
-import { ModelV2 } from "../model"
-
-export class Options extends Schema.Class<Options>("ConfigV2.Provider.Options")({
-  headers: Schema.Record(Schema.String, Schema.String).pipe(Schema.optional),
-  body: Schema.Record(Schema.String, Schema.Unknown).pipe(Schema.optional),
-  aisdk: Schema.Struct({
-    provider: Schema.Record(Schema.String, Schema.Unknown).pipe(Schema.optional),
-    request: Schema.Record(Schema.String, Schema.Unknown).pipe(Schema.optional),
-  }).pipe(Schema.optional),
-}) {}
-
-class Cache extends Schema.Class<Cache>("ConfigV2.Model.Cost.Cache")({
-  read: Schema.Finite.pipe(Schema.optional),
-  write: Schema.Finite.pipe(Schema.optional),
-}) {}
-
-class Cost extends Schema.Class<Cost>("ConfigV2.Model.Cost")({
-  tier: Schema.Struct({
-    type: Schema.Literal("context"),
-    size: Schema.Int,
-  }).pipe(Schema.optional),
-  input: Schema.Finite,
-  output: Schema.Finite,
-  cache: Cache.pipe(Schema.optional),
-}) {}
-
-class Limit extends Schema.Class<Limit>("ConfigV2.Model.Limit")({
-  context: Schema.Int.pipe(Schema.optional),
-  input: Schema.Int.pipe(Schema.optional),
-  output: Schema.Int.pipe(Schema.optional),
-}) {}
-
-class Model extends Schema.Class<Model>("ConfigV2.Model")({
-  api_id: ModelV2.ID.pipe(Schema.optional),
-  family: ModelV2.Family.pipe(Schema.optional),
-  name: Schema.String.pipe(Schema.optional),
-  endpoint: ProviderV2.Endpoint.pipe(Schema.optional),
-  capabilities: ModelV2.Capabilities.pipe(Schema.optional),
-  options: Schema.Struct({
-    ...Options.fields,
-    variant: Schema.String.pipe(Schema.optional),
-  }).pipe(Schema.optional),
-  variants: Schema.Struct({
-    id: ModelV2.VariantID,
-    ...Options.fields,
-  }).pipe(Schema.Array, Schema.optional),
-  cost: Schema.Union([Cost, Cost.pipe(Schema.Array)]).pipe(Schema.optional),
-  disabled: Schema.Boolean.pipe(Schema.optional),
-  limit: Limit.pipe(Schema.optional),
-}) {}
-
-export class Info extends Schema.Class<Info>("ConfigV2.Provider")({
-  name: Schema.String.pipe(Schema.optional),
-  env: Schema.String.pipe(Schema.Array, Schema.optional),
-  endpoint: ProviderV2.Endpoint.pipe(Schema.optional),
-  options: Options.pipe(Schema.optional),
-  models: Schema.Record(Schema.String, Model).pipe(Schema.optional),
-}) {}

packages/core/src/config/reference.ts

@@ -1,17 +0,0 @@
-export * as ConfigReference from "./reference"
-
-import { Schema } from "effect"
-
-export class Git extends Schema.Class<Git>("ConfigV2.Reference.Git")({
-  repository: Schema.String,
-  branch: Schema.String.pipe(Schema.optional),
-}) {}
-
-export class Local extends Schema.Class<Local>("ConfigV2.Reference.Local")({
-  path: Schema.String,
-}) {}
-
-export const Entry = Schema.Union([Schema.String, Git, Local])
-export type Entry = typeof Entry.Type
-
-export const Info = Schema.Record(Schema.String, Entry)

packages/core/src/config/tool-output.ts

@@ -1,9 +0,0 @@
-export * as ConfigToolOutput from "./tool-output"
-
-import { Schema } from "effect"
-import { PositiveInt } from "../schema"
-
-export class Info extends Schema.Class<Info>("ConfigV2.ToolOutput")({
-  max_lines: PositiveInt.pipe(Schema.optional),
-  max_bytes: PositiveInt.pipe(Schema.optional),
-}) {}

packages/core/src/config/watcher.ts

@@ -1,7 +0,0 @@
-export * as ConfigWatcher from "./watcher"
-
-import { Schema } from "effect"
-
-export class Info extends Schema.Class<Info>("ConfigV2.Watcher")({
-  ignore: Schema.String.pipe(Schema.Array, Schema.optional),
-}) {}

packages/core/src/event.ts

@@ -128,7 +128,7 @@ export const layer = Layer.effect(
           ...(options?.metadata ? { metadata: options.metadata } : {}),
           type: definition.type,
           ...(definition.version === undefined ? {} : { version: definition.version }),
-          ...(location ? { location: { directory: location.directory, workspaceID: location.workspaceID } } : {}),
+          ...(location ? { location } : {}),
           data,
         } as Payload<D>
         return yield* publishEvent(event)

packages/core/src/flag/flag.ts

@@ -47,7 +47,6 @@ export const Flag = {
 
   OPENCODE_WORKSPACE_ID: process.env["OPENCODE_WORKSPACE_ID"],
   OPENCODE_EXPERIMENTAL_WORKSPACES: enabledByExperimental("OPENCODE_EXPERIMENTAL_WORKSPACES"),
-  OPENCODE_EXPERIMENTAL_SESSION_SWITCHER: enabledByExperimental("OPENCODE_EXPERIMENTAL_SESSION_SWITCHER"),
 
   // Evaluated at access time (not module load) because tests, the CLI, and
   // external tooling set these env vars at runtime.

packages/core/src/location-layer.ts

@@ -2,17 +2,12 @@ import { Layer, LayerMap } from "effect"
 import { Location } from "./location"
 import { Catalog } from "./catalog"
 import { PluginBoot } from "./plugin/boot"
-import { Policy } from "./policy"
-import { Config } from "./config"
 
 export class LocationServiceMap extends LayerMap.Service<LocationServiceMap>()("@opencode/example/LocationServiceMap", {
-  lookup: (ref: Location.Ref) => {
-    const result = Layer.mergeAll(Catalog.defaultLayer, PluginBoot.defaultLayer, Config.defaultLayer).pipe(
-      Layer.provideMerge(Policy.defaultLayer),
-      Layer.provideMerge(Location.defaultLayer(ref)),
-    )
-    return result
-  },
-  idleTimeToLive: "60 minutes",
+  lookup: (ref: Location.Ref) =>
+    Layer.mergeAll(Catalog.defaultLayer, PluginBoot.defaultLayer).pipe(
+      Layer.provide([Layer.succeed(Location.Service, Location.Service.of(ref))]),
+    ),
+  idleTimeToLive: "5 minutes",
   dependencies: [],
 }) {}

packages/core/src/location.ts

@@ -1,40 +1,11 @@
-import { Context, Effect, Layer, Schema } from "effect"
-import { Project } from "./project"
-import { AbsolutePath } from "./schema"
+import { Context, Schema } from "effect"
 
 export * as Location from "./location"
 
 export const Ref = Schema.Struct({
-  directory: AbsolutePath,
+  directory: Schema.String,
   workspaceID: Schema.optional(Schema.String),
 }).annotate({ identifier: "Location.Ref" })
 export type Ref = typeof Ref.Type
 
-export interface Interface {
-  readonly directory: AbsolutePath
-  readonly workspaceID?: string
-  readonly project: {
-    readonly id: Project.ID
-    readonly directory: AbsolutePath
-  }
-  readonly vcs?: Project.Vcs
-}
-
-export class Service extends Context.Service<Service, Interface>()("@opencode/Location") {}
-
-export const layer = (ref: Ref) =>
-  Layer.effect(
-    Service,
-    Effect.gen(function* () {
-      const project = yield* Project.Service
-      const resolved = yield* project.resolve(ref.directory)
-      return Service.of({
-        directory: ref.directory,
-        workspaceID: ref.workspaceID,
-        project: { id: resolved.id, directory: resolved.directory },
-        vcs: resolved.vcs,
-      })
-    }),
-  )
-
-export const defaultLayer = (ref: Ref) => layer(ref).pipe(Layer.provide(Project.defaultLayer))
+export class Service extends Context.Service<Service, Ref>()("@opencode/Location") {}

packages/core/src/model.ts

@@ -36,7 +36,7 @@ export const Cost = Schema.Struct({
 export const Ref = Schema.Struct({
   id: ID,
   providerID: ProviderV2.ID,
-  variant: VariantID.pipe(Schema.optional),
+  variant: VariantID,
 })
 export type Ref = typeof Ref.Type
 

packages/core/src/plugin.ts

@@ -2,26 +2,17 @@ export * as PluginV2 from "./plugin"
 
 import { createDraft, finishDraft, type Draft } from "immer"
 import type { LanguageModelV3 } from "@ai-sdk/provider"
-import { Context, Effect, Exit, Layer, Schema, Scope } from "effect"
+import { Context, Effect, Exit, Layer, PubSub, Schema, Scope, Stream } from "effect"
 import type { ModelV2 } from "./model"
+import type { AgentV2 } from "./agent"
 import type { Catalog } from "./catalog"
-import { EventV2 } from "./event"
 
 export const ID = Schema.String.pipe(Schema.brand("Plugin.ID"))
 export type ID = typeof ID.Type
 
-export const Event = {
-  Added: EventV2.define({
-    type: "plugin.added",
-    schema: {
-      id: ID,
-    },
-  }),
-}
-
 type HookSpec = {
   "catalog.transform": {
-    input: Catalog.Editor
+    input: Catalog.Context
     output: {}
   }
   "account.switched": {
@@ -52,6 +43,27 @@ type HookSpec = {
       sdk?: any
     }
   }
+  "agent.update": {
+    input: {}
+    output: {
+      agent: AgentV2.Info
+      cancel: boolean
+    }
+  }
+  "agent.remove": {
+    input: {
+      agent: AgentV2.Info
+    }
+    output: {
+      cancel: boolean
+    }
+  }
+  "agent.default": {
+    input: {}
+    output: {
+      agent?: AgentV2.ID
+    }
+  }
 }
 
 export type Hooks = {
@@ -81,6 +93,7 @@ export interface Interface {
     effect: Effect.Effect<void | HookFunctions, never, Scope.Scope>
   }) => Effect.Effect<void, never, never>
   readonly remove: (id: ID) => Effect.Effect<void>
+  readonly added: () => Stream.Stream<ID>
   readonly triggerFor: <Name extends keyof Hooks>(
     id: ID,
     name: Name,
@@ -104,7 +117,9 @@ export const layer = Layer.effect(
       hooks: HookFunctions
       scope: Scope.Closeable
     }[] = []
-    const events = yield* EventV2.Service
+    const added = yield* PubSub.unbounded<ID>()
+
+    yield* Effect.addFinalizer(() => PubSub.shutdown(added))
 
     const svc = Service.of({
       add: Effect.fn("Plugin.add")(function* (input) {
@@ -120,8 +135,9 @@ export const layer = Layer.effect(
             scope,
           },
         ]
-        yield* events.publish(Event.Added, { id: input.id })
+        yield* PubSub.publish(added, input.id)
       }),
+      added: () => Stream.fromPubSub(added),
       trigger: Effect.fn("Plugin.trigger")(function* (name, input, output) {
         return yield* svc.triggerFor(ID.make("*"), name, input, output)
       }),
@@ -169,7 +185,7 @@ export const layer = Layer.effect(
   }),
 )
 
-export const defaultLayer = layer.pipe(Layer.provide(EventV2.defaultLayer))
+export const defaultLayer = layer
 
 // opencode
 // sdcok

packages/core/src/plugin/account.ts

@@ -19,7 +19,7 @@ export const AccountPlugin = PluginV2.define({
 
     return {
       "catalog.transform": Effect.fn(function* (evt) {
-        for (const item of evt.provider.list()) {
+        for (const item of evt.data) {
           const account = yield* accounts.active(AccountV2.ServiceID.make(item.provider.id)).pipe(Effect.orDie)
           if (!account) continue
           evt.provider.update(item.provider.id, (provider) => {

packages/core/src/plugin/agent.ts

@@ -1,211 +0,0 @@
-export * as AgentPlugin from "./agent"
-
-import path from "path"
-import { Effect } from "effect"
-import { AgentV2 } from "../agent"
-import { Global } from "../global"
-import { Location } from "../location"
-import { PermissionV2 } from "../permission"
-import { PluginV2 } from "../plugin"
-
-const TRUNCATION_GLOB = path.join(Global.Path.data, "tool-output", "*")
-
-const PROMPT_EXPLORE = `You are a file search specialist. You excel at thoroughly navigating and exploring codebases.
-
-Your strengths:
-- Rapidly finding files using glob patterns
-- Searching code and text with powerful regex patterns
-- Reading and analyzing file contents
-
-Guidelines:
-- Use Glob for broad file pattern matching
-- Use Grep for searching file contents with regex
-- Use Read when you know the specific file path you need to read
-- Use Bash for file operations like copying, moving, or listing directory contents
-- Adapt your search approach based on the thoroughness level specified by the caller
-- Return file paths as absolute paths in your final response
-- For clear communication, avoid using emojis
-- Do not create any files, or run bash commands that modify the user's system state in any way
-
-Complete the user's search request efficiently and report your findings clearly.`
-
-const PROMPT_COMPACTION = `You are an anchored context summarization assistant for coding sessions.
-
-Summarize only the conversation history you are given. The newest turns may be kept verbatim outside your summary, so focus on the older context that still matters for continuing the work.
-
-If the prompt includes a <previous-summary> block, treat it as the current anchored summary. Update it with the new history by preserving still-true details, removing stale details, and merging in new facts.
-
-Always follow the exact output structure requested by the user prompt. Keep every section, preserve exact file paths and identifiers when known, and prefer terse bullets over paragraphs.
-
-Do not answer the conversation itself. Do not mention that you are summarizing, compacting, or merging context. Respond in the same language as the conversation.`
-
-const PROMPT_TITLE = `You are a title generator. You output ONLY a thread title. Nothing else.
-
-<task>
-Generate a brief title that would help the user find this conversation later.
-
-Follow all rules in <rules>
-Use the <examples> so you know what a good title looks like.
-Your output must be:
-- A single line
-- <=50 characters
-- No explanations
-</task>
-
-<rules>
-- you MUST use the same language as the user message you are summarizing
-- Title must be grammatically correct and read naturally - no word salad
-- Never include tool names in the title (e.g. "read tool", "bash tool", "edit tool")
-- Focus on the main topic or question the user needs to retrieve
-- Vary your phrasing - avoid repetitive patterns like always starting with "Analyzing"
-- When a file is mentioned, focus on WHAT the user wants to do WITH the file, not just that they shared it
-- Keep exact: technical terms, numbers, filenames, HTTP codes
-- Remove: the, this, my, a, an
-- Never assume tech stack
-- Never use tools
-- NEVER respond to questions, just generate a title for the conversation
-- The title should NEVER include "summarizing" or "generating" when generating a title
-- DO NOT SAY YOU CANNOT GENERATE A TITLE OR COMPLAIN ABOUT THE INPUT
-- Always output something meaningful, even if the input is minimal.
-- If the user message is short or conversational (e.g. "hello", "lol", "what's up", "hey"):
-  -> create a title that reflects the user's tone or intent (such as Greeting, Quick check-in, Light chat, Intro message, etc.)
-</rules>
-
-<examples>
-"debug 500 errors in production" -> Debugging production 500 errors
-"refactor user service" -> Refactoring user service
-"why is app.js failing" -> app.js failure investigation
-"implement rate limiting" -> Rate limiting implementation
-"how do I connect postgres to my API" -> Postgres API connection
-"best practices for React hooks" -> React hooks best practices
-"@src/auth.ts can you add refresh token support" -> Auth refresh token support
-"@utils/parser.ts this is broken" -> Parser bug fix
-"look at @config.json" -> Config review
-"@App.tsx add dark mode toggle" -> Dark mode toggle in App
-</examples>`
-
-const PROMPT_SUMMARY = `Summarize what was done in this conversation. Write like a pull request description.
-
-Rules:
-- 2-3 sentences max
-- Describe the changes made, not the process
-- Do not mention running tests, builds, or other validation steps
-- Do not explain what the user asked for
-- Write in first person (I added..., I fixed...)
-- Never ask questions or add new questions
-- If the conversation ends with an unanswered question to the user, preserve that exact question
-- If the conversation ends with an imperative statement or request to the user (e.g. "Now please run the command and paste the console output"), always include that exact request in the summary`
-
-export const Plugin = PluginV2.define({
-  id: PluginV2.ID.make("agent"),
-  effect: Effect.gen(function* () {
-    const agent = yield* AgentV2.Service
-    const location = yield* Location.Service
-    const worktree = location.directory
-    const whitelistedDirs = [TRUNCATION_GLOB, path.join(Global.Path.tmp, "*")]
-    const readonlyExternalDirectory: PermissionV2.Ruleset = [
-      { permission: "external_directory", pattern: "*", action: "ask" },
-      ...whitelistedDirs.map(
-        (pattern): PermissionV2.Rule => ({ permission: "external_directory", pattern, action: "allow" }),
-      ),
-    ]
-    const defaults: PermissionV2.Ruleset = [
-      { permission: "*", pattern: "*", action: "allow" },
-      ...readonlyExternalDirectory,
-      { permission: "question", pattern: "*", action: "deny" },
-      { permission: "plan_enter", pattern: "*", action: "deny" },
-      { permission: "plan_exit", pattern: "*", action: "deny" },
-      { permission: "repo_clone", pattern: "*", action: "deny" },
-      { permission: "repo_overview", pattern: "*", action: "deny" },
-      { permission: "read", pattern: "*", action: "allow" },
-      { permission: "read", pattern: "*.env", action: "ask" },
-      { permission: "read", pattern: "*.env.*", action: "ask" },
-      { permission: "read", pattern: "*.env.example", action: "allow" },
-    ]
-
-    yield* agent.update((editor) => {
-      editor.update(AgentV2.ID.make("build"), (item) => {
-        item.description = "The default agent. Executes tools based on configured permissions."
-        item.mode = "primary"
-        item.permissions.push(
-          ...PermissionV2.merge(defaults, [
-            { permission: "question", pattern: "*", action: "allow" },
-            { permission: "plan_enter", pattern: "*", action: "allow" },
-          ]),
-        )
-      })
-
-      editor.update(AgentV2.ID.make("plan"), (item) => {
-        item.description = "Plan mode. Disallows all edit tools."
-        item.mode = "primary"
-        item.permissions.push(
-          ...PermissionV2.merge(defaults, [
-            { permission: "question", pattern: "*", action: "allow" },
-            { permission: "plan_exit", pattern: "*", action: "allow" },
-            { permission: "external_directory", pattern: path.join(Global.Path.data, "plans", "*"), action: "allow" },
-            { permission: "edit", pattern: "*", action: "deny" },
-            { permission: "edit", pattern: path.join(".opencode", "plans", "*.md"), action: "allow" },
-            {
-              permission: "edit",
-              pattern: path.relative(worktree, path.join(Global.Path.data, "plans", "*.md")),
-              action: "allow",
-            },
-          ]),
-        )
-      })
-
-      editor.update(AgentV2.ID.make("general"), (item) => {
-        item.description =
-          "General-purpose agent for researching complex questions and executing multi-step tasks. Use this agent to execute multiple units of work in parallel."
-        item.mode = "subagent"
-        item.permissions.push(
-          ...PermissionV2.merge(defaults, [{ permission: "todowrite", pattern: "*", action: "deny" }]),
-        )
-      })
-
-      editor.update(AgentV2.ID.make("explore"), (item) => {
-        item.description =
-          'Fast agent specialized for exploring codebases. Use this when you need to quickly find files by patterns (eg. "src/components/**/*.tsx"), search code for keywords (eg. "API endpoints"), or answer questions about the codebase (eg. "how do API endpoints work?"). When calling this agent, specify the desired thoroughness level: "quick" for basic searches, "medium" for moderate exploration, or "very thorough" for comprehensive analysis across multiple locations and naming conventions.'
-        item.system = PROMPT_EXPLORE
-        item.mode = "subagent"
-        item.permissions.push(
-          ...PermissionV2.merge(
-            defaults,
-            [
-              { permission: "*", pattern: "*", action: "deny" },
-              { permission: "grep", pattern: "*", action: "allow" },
-              { permission: "glob", pattern: "*", action: "allow" },
-              { permission: "list", pattern: "*", action: "allow" },
-              { permission: "bash", pattern: "*", action: "allow" },
-              { permission: "webfetch", pattern: "*", action: "allow" },
-              { permission: "websearch", pattern: "*", action: "allow" },
-              { permission: "read", pattern: "*", action: "allow" },
-            ],
-            readonlyExternalDirectory,
-          ),
-        )
-      })
-
-      editor.update(AgentV2.ID.make("compaction"), (item) => {
-        item.mode = "primary"
-        item.hidden = true
-        item.system = PROMPT_COMPACTION
-        item.permissions.push(...PermissionV2.merge(defaults, [{ permission: "*", pattern: "*", action: "deny" }]))
-      })
-
-      editor.update(AgentV2.ID.make("title"), (item) => {
-        item.mode = "primary"
-        item.hidden = true
-        item.system = PROMPT_TITLE
-        item.permissions.push(...PermissionV2.merge(defaults, [{ permission: "*", pattern: "*", action: "deny" }]))
-      })
-
-      editor.update(AgentV2.ID.make("summary"), (item) => {
-        item.mode = "primary"
-        item.hidden = true
-        item.system = PROMPT_SUMMARY
-        item.permissions.push(...PermissionV2.merge(defaults, [{ permission: "*", pattern: "*", action: "deny" }]))
-      })
-    })
-  }),
-})

packages/core/src/plugin/boot.ts

@@ -4,15 +4,10 @@ import { Context, Deferred, Effect, Layer } from "effect"
 import { AccountV2 } from "../account"
 import { AgentV2 } from "../agent"
 import { Catalog } from "../catalog"
-import { Config } from "../config"
-import { ConfigAgentPlugin } from "../config/plugin/agent"
 import { EventV2 } from "../event"
-import { Location } from "../location"
 import { Npm } from "../npm"
 import { PluginV2 } from "../plugin"
 import { AccountPlugin } from "./account"
-import { AgentPlugin } from "./agent"
-import { ConfigProviderPlugin } from "../config/plugin/provider"
 import { EnvPlugin } from "./env"
 import { ModelsDevPlugin } from "./models-dev"
 import { ProviderPlugins } from "./provider"
@@ -20,14 +15,7 @@ import { ProviderPlugins } from "./provider"
 type Plugin = {
   id: PluginV2.ID
   effect: PluginV2.Effect<
-    | Catalog.Service
-    | AccountV2.Service
-    | AgentV2.Service
-    | Npm.Service
-    | EventV2.Service
-    | Location.Service
-    | PluginV2.Service
-    | Config.Service
+    Catalog.Service | AgentV2.Service | AccountV2.Service | Npm.Service | EventV2.Service | PluginV2.Service
   >
 }
 
@@ -40,12 +28,10 @@ export class Service extends Context.Service<Service, Interface>()("@opencode/v2
 export const layer = Layer.effect(
   Service,
   Effect.gen(function* () {
+    const agent = yield* AgentV2.Service
     const catalog = yield* Catalog.Service
     const plugin = yield* PluginV2.Service
     const accounts = yield* AccountV2.Service
-    const agents = yield* AgentV2.Service
-    const config = yield* Config.Service
-    const location = yield* Location.Service
     const npm = yield* Npm.Service
     const events = yield* EventV2.Service
     const done = yield* Deferred.make<void>()
@@ -55,10 +41,8 @@ export const layer = Layer.effect(
         id: input.id,
         effect: input.effect.pipe(
           Effect.provideService(Catalog.Service, catalog),
+          Effect.provideService(AgentV2.Service, agent),
           Effect.provideService(AccountV2.Service, accounts),
-          Effect.provideService(AgentV2.Service, agents),
-          Effect.provideService(Config.Service, config),
-          Effect.provideService(Location.Service, location),
           Effect.provideService(Npm.Service, npm),
           Effect.provideService(EventV2.Service, events),
           Effect.provideService(PluginV2.Service, plugin),
@@ -69,13 +53,10 @@ export const layer = Layer.effect(
     const boot = Effect.gen(function* () {
       yield* add(EnvPlugin)
       yield* add(AccountPlugin)
-      yield* add(AgentPlugin.Plugin)
       for (const item of ProviderPlugins) {
         yield* add(item)
       }
       yield* add(ModelsDevPlugin)
-      yield* add(ConfigProviderPlugin.Plugin)
-      yield* add(ConfigAgentPlugin.Plugin)
     }).pipe(Effect.withSpan("PluginBoot.boot"))
 
     yield* boot.pipe(
@@ -91,11 +72,10 @@ export const layer = Layer.effect(
 )
 
 export const defaultLayer = layer.pipe(
+  Layer.provide(AgentV2.defaultLayer),
   Layer.provide(Catalog.defaultLayer),
   Layer.provide(EventV2.defaultLayer),
   Layer.provide(PluginV2.defaultLayer),
   Layer.provide(AccountV2.defaultLayer),
-  Layer.provide(AgentV2.defaultLayer),
-  Layer.provide(Config.defaultLayer),
   Layer.provide(Npm.defaultLayer),
 )

packages/core/src/plugin/env.ts

@@ -6,7 +6,7 @@ export const EnvPlugin = PluginV2.define({
   effect: Effect.gen(function* () {
     return {
       "catalog.transform": Effect.fn(function* (evt) {
-        for (const item of evt.provider.list()) {
+        for (const item of evt.data) {
           const key = item.provider.env.find((env) => process.env[env])
           if (!key) continue
           evt.provider.update(item.provider.id, (provider) => {

packages/core/src/plugin/models-dev.ts

@@ -57,10 +57,10 @@ export const ModelsDevPlugin = PluginV2.define({
     const modelsDev = yield* ModelsDev.Service
     const events = yield* EventV2.Service
     const scope = yield* Scope.Scope
-    const transform = yield* catalog.transform()
+    const load = yield* catalog.loader()
     const refresh = Effect.fn("ModelsDevPlugin.refresh")(function* () {
       const data = yield* modelsDev.get()
-      yield* transform((catalog) => {
+      yield* load((catalog) => {
         for (const item of Object.values(data)) {
           const providerID = ProviderV2.ID.make(item.id)
           catalog.provider.update(providerID, (provider) => {

packages/core/src/plugin/provider/amazon-bedrock.ts

@@ -51,7 +51,7 @@ export const AmazonBedrockPlugin = PluginV2.define({
   effect: Effect.gen(function* () {
     return {
       "catalog.transform": Effect.fn(function* (evt) {
-        for (const item of evt.provider.list()) {
+        for (const item of evt.data) {
           if (item.provider.endpoint.type !== "aisdk") continue
           if (item.provider.endpoint.package !== "@ai-sdk/amazon-bedrock") continue
           evt.provider.update(item.provider.id, (provider) => {

packages/core/src/plugin/provider/anthropic.ts

@@ -6,7 +6,7 @@ export const AnthropicPlugin = PluginV2.define({
   effect: Effect.gen(function* () {
     return {
       "catalog.transform": Effect.fn(function* (evt) {
-        for (const item of evt.provider.list()) {
+        for (const item of evt.data) {
           if (item.provider.endpoint.type !== "aisdk") continue
           if (item.provider.endpoint.package !== "@ai-sdk/anthropic") continue
           evt.provider.update(item.provider.id, (provider) => {

packages/core/src/plugin/provider/azure.ts

@@ -15,7 +15,7 @@ export const AzurePlugin = PluginV2.define({
   effect: Effect.gen(function* () {
     return {
       "catalog.transform": Effect.fn(function* (evt) {
-        for (const item of evt.provider.list()) {
+        for (const item of evt.data) {
           if (item.provider.endpoint.type !== "aisdk") continue
           if (item.provider.endpoint.package !== "@ai-sdk/azure") continue
           const configured = item.provider.options.aisdk.provider.resourceName
@@ -58,7 +58,7 @@ export const AzureCognitiveServicesPlugin = PluginV2.define({
       "catalog.transform": Effect.fn(function* (evt) {
         const resourceName = process.env.AZURE_COGNITIVE_SERVICES_RESOURCE_NAME
         if (!resourceName) return
-        for (const item of evt.provider.list()) {
+        for (const item of evt.data) {
           if (item.provider.endpoint.type !== "aisdk") continue
           if (item.provider.endpoint.package !== "@ai-sdk/openai-compatible") continue
           if (!item.provider.id.includes("azure-cognitive-services")) continue

packages/core/src/plugin/provider/cerebras.ts

@@ -6,7 +6,7 @@ export const CerebrasPlugin = PluginV2.define({
   effect: Effect.gen(function* () {
     return {
       "catalog.transform": Effect.fn(function* (ctx) {
-        for (const item of ctx.provider.list()) {
+        for (const item of ctx.data) {
           if (item.provider.endpoint.type !== "aisdk") continue
           if (item.provider.endpoint.package !== "@ai-sdk/cerebras") continue
           ctx.provider.update(item.provider.id, (provider) => {

packages/core/src/plugin/provider/cloudflare-workers-ai.ts

@@ -11,7 +11,7 @@ export const CloudflareWorkersAIPlugin = PluginV2.define({
   effect: Effect.gen(function* () {
     return {
       "catalog.transform": Effect.fn(function* (evt) {
-        const item = evt.provider.get(providerID)
+        const item = evt.data.find((record) => record.provider.id === providerID)
         if (!item) return
         evt.provider.update(item.provider.id, (provider) => {
           if (provider.endpoint.type !== "aisdk") return

packages/core/src/plugin/provider/github-copilot.ts

@@ -31,7 +31,7 @@ export const GithubCopilotPlugin = PluginV2.define({
           : evt.sdk.chat(evt.model.apiID)
       }),
       "catalog.transform": Effect.fn(function* (evt) {
-        const item = evt.provider.get(ProviderV2.ID.githubCopilot)
+        const item = evt.data.find((record) => record.provider.id === ProviderV2.ID.githubCopilot)
         if (!item || !item.models.has(ModelV2.ID.make("gpt-5-chat-latest"))) return
         evt.model.update(item.provider.id, ModelV2.ID.make("gpt-5-chat-latest"), (model) => {
           // This chat-only alias conflicts with the Copilot GPT-5 Responses route,

packages/core/src/plugin/provider/google-vertex.ts

@@ -43,9 +43,9 @@ function authFetch(fetchWithRuntimeOptions?: unknown) {
   // do not, so inject a Google access token into their fetch path.
   return async (input: Parameters<typeof fetch>[0], init?: RequestInit) => {
     const { GoogleAuth } = await import("google-auth-library")
-    const auth = new GoogleAuth({ scopes: ["https://www.googleapis.com/auth/cloud-platform"] })
-    const client = await auth.getClient()
-    const token = await client.getAccessToken()
+    const auth = new GoogleAuth()
+    const client = await auth.getApplicationDefault()
+    const token = await client.credential.getAccessToken()
     const headers = new Headers(init?.headers)
     headers.set("Authorization", `Bearer ${token.token}`)
     return typeof fetchWithRuntimeOptions === "function"
@@ -59,7 +59,7 @@ export const GoogleVertexPlugin = PluginV2.define({
   effect: Effect.gen(function* () {
     return {
       "catalog.transform": Effect.fn(function* (evt) {
-        for (const item of evt.provider.list()) {
+        for (const item of evt.data) {
           if (item.provider.endpoint.type !== "aisdk") continue
           if (
             item.provider.endpoint.package !== "@ai-sdk/google-vertex" &&
@@ -110,7 +110,7 @@ export const GoogleVertexAnthropicPlugin = PluginV2.define({
   effect: Effect.gen(function* () {
     return {
       "catalog.transform": Effect.fn(function* (evt) {
-        for (const item of evt.provider.list()) {
+        for (const item of evt.data) {
           if (item.provider.endpoint.type !== "aisdk") continue
           if (item.provider.endpoint.package !== "@ai-sdk/google-vertex/anthropic") continue
           const project =

packages/core/src/plugin/provider/kilo.ts

@@ -6,7 +6,7 @@ export const KiloPlugin = PluginV2.define({
   effect: Effect.gen(function* () {
     return {
       "catalog.transform": Effect.fn(function* (evt) {
-        for (const item of evt.provider.list()) {
+        for (const item of evt.data) {
           if (item.provider.endpoint.type !== "aisdk") continue
           if (item.provider.endpoint.package !== "@ai-sdk/openai-compatible") continue
           if (item.provider.endpoint.url !== "https://api.kilo.ai/api/gateway") continue

packages/core/src/plugin/provider/llmgateway.ts

@@ -6,7 +6,7 @@ export const LLMGatewayPlugin = PluginV2.define({
   effect: Effect.gen(function* () {
     return {
       "catalog.transform": Effect.fn(function* (evt) {
-        for (const item of evt.provider.list()) {
+        for (const item of evt.data) {
           if (item.provider.enabled === false) continue
           if (item.provider.endpoint.type !== "aisdk") continue
           if (item.provider.endpoint.package !== "@ai-sdk/openai-compatible") continue