chore: sync upstream/main (0.36.0) into fork/main

.agents/skills/agent-core-dev/config.md

@@ -101,7 +101,7 @@ pass `ConfigTarget.Memory` for a per-run override that is never written to disk.
 - `src/kosong/model/thinking.ts` (owner domain, not `config`) — the `resolveThinkingEffort` helper and the authoritative `ThinkingConfig` type (the `thinking` section itself registers from `src/app/kosongConfig/configSection.ts`).
 - `src/app/config/configPure.ts` — `isPlainObject`, `deepMerge`, `omitUndefined`, `describeUnknownError`.
 
-A domain that owns a section keeps the schema in its own `configSection.ts` (e.g. `src/app/flag/flag.ts` for `experimental`, `src/agent/loop/configSection.ts` for `loopControl`). Exception: kosong-owned sections (`providers`, `models`, `thinking`) — kosong is a pure, persistence-free abstraction layer that defines only the types (`src/kosong/{provider,model}`); the section constants, the zod schemas (re-derived from those types and compile-time pinned via `AssertExact<Equal<z.infer<typeof Schema>, Type>>`, see `_base/utils/typeEquality.ts`), the registrations, env bindings, and TOML transforms all live in the persistence wrapper `src/app/kosongConfig/configSection.ts`. (`modelCatalog` and `secondaryModel` have no kosong-side type at all — their sections are fully self-contained in `app/kosongConfig`, types derived from the schemas.) A cross-section env overlay (e.g. the `KIMI_MODEL_*` synthesis) lives in the wrapper too (`src/app/kosongConfig/envOverlay.ts`; the `[secondary_model]` derived-entry synthesis in `secondaryModelOverlay.ts`) and is registered via module-level `registerConfigOverlay`. The two-way sync between config sections and kosong's in-memory registries is owned by `IKosongConfigService` (`src/app/kosongConfig/kosongConfigService.ts`).
+A domain that owns a section keeps the schema in its own `configSection.ts` (e.g. `src/app/flag/flag.ts` for `experimental`, `src/agent/loop/configSection.ts` for `loopControl`). Exception: kosong-owned sections (`providers`, `models`, `thinking`) — kosong is a pure, persistence-free abstraction layer that defines only the types (`src/kosong/{provider,model}`); the section constants, the zod schemas (re-derived from those types and compile-time pinned via `AssertExact<Equal<z.infer<typeof Schema>, Type>>`, see `_base/utils/typeEquality.ts`), the registrations, env bindings, and TOML transforms all live in the persistence wrapper `src/app/kosongConfig/configSection.ts`. (`modelCatalog` has no kosong-side type at all — its section is fully self-contained in `app/kosongConfig`, types derived from the schema.) A cross-section env overlay (e.g. the `KIMI_MODEL_*` synthesis in `src/app/kosongConfig/envOverlay.ts`) lives in the wrapper too and is registered via module-level `registerConfigOverlay`. The session subagent domain owns two sections in `src/session/subagent/configSection.ts`: `[subagent]` (`timeout_ms` on disk) and `[secondary_model]` (`default_model` plus the `[secondary_model.models]` pool, with a lone legacy v1 `model` key honored as a fallback default below `default_model`); neither carries a cross-section overlay. Cross-field pool validation (default present / in-pool / every key resolvable) runs at session creation in `subagentModelsValidationService.ts`, not in the schema. The two-way sync between config sections and kosong's in-memory registries is owned by `IKosongConfigService` (`src/app/kosongConfig/kosongConfigService.ts`).
 
 ## Scope
 

.agents/skills/agent-core-dev/edge-exposure.md

@@ -45,7 +45,7 @@ A Service method is directly exposable iff **all** hold:
 3. Errors are `KimiError` (coded).
 4. It is a command/query, not a factory, stream, byte-store, or sink.
 
-If any fail → wrap in a **facade** (a Service that takes ids, returns data, throws `KimiError`) and expose the facade. The repo already ships a wire-shaped facade in `rpc/core-api.ts` (`CoreAPI` / `SessionAPI` / `AgentAPI`) behind `IAgentRPCService` / `ISessionRPCService` — prefer building the HTTP edge on top of it rather than re-deriving a new one.
+If any fail → add a wire-safe orchestration method to the owning domain Service (e.g. `IAgentPromptService.submit` settles `{turn_id}` instead of returning the live `PromptHandle`) or compose several domain Services at the edge — kap-server's `routes/prompts.ts` is the reference for edge-side composition.
 
 ## 3. Per-scope `resource:action` map
 

.agents/skills/agent-core-dev/server-align.md

@@ -165,7 +165,7 @@ const route = defineRoute(
 app.post(route.path, route.options, route.handler);
 ```
 
-**For `/api/v2` (native):** add a `resource:action` entry to `actionMap` ([edge-exposure.md](edge-exposure.md) §3). If the method fails the direct-exposure rules (returns a handle / stream / bytes, takes a live object), wrap it in a wire-shaped facade first (`IAgentRPCService` / `ISessionRPCService`) and map to the facade — as `prompts:*` does via `IAgentRPCService`.
+**For `/api/v2` (native):** add a `resource:action` entry to `actionMap` ([edge-exposure.md](edge-exposure.md) §3). If the method fails the direct-exposure rules (returns a handle / stream / bytes, takes a live object), add a wire-safe orchestration method to the owning domain Service first — as `prompts:submit` maps to `IAgentPromptService.submit`, which settles `{turn_id}` engine-side instead of returning the live `PromptHandle`.
 
 ### 5. Map errors
 
@@ -218,7 +218,7 @@ This is the reference alignment (commits `feat(server-v2): port v1 /sessions/:si
 
 **The split.**
 
-- `/api/v2` keeps the native shape — `prompts:submit` / `steer` / `undo` / `clear` / `cancel` map to `IAgentRPCService` (a wire facade over the v2 turn driver) in `actionMap`. The native `IAgentPromptService` is untouched.
+- `/api/v2` keeps the native shape — `prompts:submit` / `steer` / `undo` / `clear` / `cancel` map to the domain Services (`IAgentPromptService.submit` / `submitSteer`, `IAgentConversationUndoService.undo`, `IAgentLoopService.cancelFromUser`) in `actionMap`.
 - `/api/v1` gets an `AgentPromptLegacyService` (`prompt/`, `LifecycleScope.Agent`) that re-implements the v1 scheduler — queue, `prompt_id`, steer/abort, auto-start-next — **on top of** the native `IAgentPromptService`. The `/api/v1` routes consume the LegacyService.
 
 **The schema.** Both surfaces import `promptSubmissionSchema` / `promptSubmitResultSchema` / `promptListResponseSchema` / `promptSteerRequestSchema` / `promptSteerResultSchema` / `promptAbortResponseSchema` from the shared v1 wire schemas (see `packages/kap-server/src/protocol`). The `/api/v1` and `/api/v2` routes are therefore compatible with released clients by construction; the LegacyService projects v2 turn results back into those protocol shapes.

.agents/skills/agent-core-dev/service-authoring.md

@@ -51,7 +51,7 @@ File names derive from the interface / class names so that scope and role are vi
 | Shared-types file | `<domain>.types.ts` | `log.types.ts` |
 | Errors file | `<name>.errors.ts` | `appendLogStore.errors.ts` |
 
-Acronym-aware lowerCamelCase lowercases a leading acronym as a group: `ILLMRequester` → `llmRequester.ts`, `IWSGateway` → `wsGateway.ts`, `IOAuthToolkit` → `oauthToolkit.ts`, `IAgentRPCService` → `agentRpcService.ts`.
+Acronym-aware lowerCamelCase lowercases a leading acronym as a group: `ILLMRequester` → `llmRequester.ts`, `IWSGateway` → `wsGateway.ts`, `IOAuthToolkit` → `oauthToolkit.ts`, `IMcpServerService` → `mcpServerService.ts`.
 
 Because the impl class always ends in `Service` and the interface file never does, the two files of one service never collide — even for `Store` / `Registry` / `Resolver` interfaces (`IAppendLogStore` → `appendLogStore.ts` + `appendLogStoreService.ts`).
 

.changeset/agent-core-v2-session-title.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/agent-core-v2": minor
+---
+
+Add the Session-scoped `ISessionTitleService` for managed AI session titles: composes the excerpt sent to the platform chat_title tool from the main agent's conversation (the first user prompts, the strict `first_turn` pair, or the head+tail `digest` for multi-turn sessions; assistant segments keep only final text), persists the result with a `titleKind` (`replaceable` / `generated` / `custom`) that never overwrites a user-renamed title unless explicitly forced, and rebroadcasts `session.meta.updated`. Gated by the new experimental `auto_session_title` flag and a managed OAuth login.

.changeset/auto-session-title.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kimi-code-sdk": minor
+---
+
+Add `generateSessionTitle` (v2 engine) for managed AI session titles: optional `force` regeneration over generated/custom titles and selectable conversation excerpts (`user_prompts` / `first_turn` / `digest`). Gated by the experimental `auto_session_title` flag and a managed OAuth login.

.changeset/fix-banner-long-tag-narrow-terminal.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kimi-code": patch
+---
+
+Fix startup banner text wrapping on narrow terminals.

.changeset/fix-mcp-oauth-cancel.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kimi-code": patch
+---
+
+Fix MCP OAuth cancellation leaving an in-flight authorization waiting for its callback timeout.

.changeset/fix-windows-project-root-watch.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kimi-code": patch
+---
+
+Fix repeated file-watcher errors on Windows when the workspace is a drive root (such as `E:\`) or a UNC network share.

.changeset/kap-server-session-title-route.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kap-server": patch
+---
+
+Add `POST /api/v1/sessions/{session_id}/title/generate` with an optional `{ "force": true, "source": "user_prompts" | "first_turn" | "digest" }` body; unknown sessions return 40401 and unavailable generation (flag off, no managed login, no prompt yet, backend failure) returns the new 40923 SESSION_TITLE_UNAVAILABLE.

.changeset/klient-session-title.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/klient": patch
+---
+
+Expose `session(id).generateTitle({ force, source })` on the session facade and the matching `sessionTitleService` wire contract.

.changeset/oauth-chat-title.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kimi-code-oauth": patch
+---
+
+Add `fetchChatTitle` for the managed platform `/tools` `chat_title` method: protocol headers, an 8s timeout, response validation, and structured failures.

.changeset/oauth-device-subpath-export.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kimi-code-oauth": minor
+---
+
+Add a browser-safe `./device` subpath export exposing the device-code flow's pure-fetch HTTP wrappers and flow config, so browser bundles can run OAuth sign-in without pulling in Node-only modules. Import from `@moonshot-ai/kimi-code-oauth/device`.

apps/kimi-code/AGENTS.md

@@ -65,6 +65,7 @@ The theme apply/switch mechanics live in the `write-tui` skill. The following ru
 
 ## General Coding Requirements
 
+- The startup path before the workspace trust gate (`KimiTUI.start()` -> `maybeRunWorkspaceTrustPrompt()`) must not spawn child processes by bare command name — on Windows, cmd.exe / CreateProcess resolve them from the current directory first, so a binary planted in an untrusted workspace would run before the user confirms trust. When an external command is unavoidable, resolve it with `resolveCommandPath` from `src/utils/process/resolve-command.ts`, which returns an absolute PATH hit and refuses matches inside the cwd.
 - For optional object properties, pass `undefined` directly — do not use conditional spread.
 - Optional object properties do not need to additionally allow `undefined` in the type.
 - Internal methods with only a single parameter should not be turned into options objects just for stylistic uniformity.