fix(tools): state the code-mode collapse in the prompt and the denial

The executor collapse landed without telling the model it exists. Every
tool contributes its own guidance section naming its tool, none of them
qualify how that tool is reached, and they all render before the SDK
(orders 100-199 against SDK_SECTION_ORDER 150), so the prompt said "Use
the read tool" eleven times and never said only run_code is callable.

A real session shows the consequence: the model emitted a native call,
read `unknown tool "read"` for a tool the same prompt declares, and
concluded the deployment was inconsistent rather than routing through
run_code.

The registry now contributes `tools:code-only` at order 99 -- ahead of
the guidance band -- stating the rule, registered wherever `tools:sdk`
is and rendering empty outside an effective `code`. `both` renders it
empty because its native calls do execute, which is also why
both-mode-turn no longer shares code-mode-turn's expected prompt. The
denial itself now names the route back, since a bare UNKNOWN_TOOL for a
declared tool is what misled the model.

.agents/notes/implemented/bug-fix/2026-08-07-code-mode-executor-collapse.md

@@ -12,7 +12,7 @@ The package contract names this exact anti-pattern: schema omission is not enfor
 
 ## Decision
 
-`ToolRegistry` resolves callable definitions through a new private `resolveExecution(name, scope, nested)` that applies the mode collapse at the operation boundary that owns it. A model-direct call (`nested = false`) under `code` may only name the reserved `run_code` transport; every native name resolves to `undefined` and surfaces as the executor's existing `UNKNOWN_TOOL` error (an already-aborted caller signal keeps the cancellation contract: `ABORTED_BEFORE_DISPATCH`, with the visible tool's finalizer applied). A collapsed call terminates at `createExecution` — the first stage of `prepare` — BEFORE the extensible policy pipeline, so `tools/pre-execute` listeners, approval `ask`, and guards never observe a call that is deterministically denied; a human is never prompted to approve it. A nested sub-dispatch (`nested = true` — a `parent` token set, which only the `run_code` SDK binding sets in production code) may call any visible tool, so programs keep every binding the generated SDK declared.
+`ToolRegistry` resolves callable definitions through a new private `resolveExecution(name, scope, nested)` that applies the mode collapse at the operation boundary that owns it. When `modeFor(scope)` resolves to `code`, a model-direct call (`nested = false`) may only name the reserved `run_code` transport; every native name resolves to `undefined` and surfaces as the executor's existing `UNKNOWN_TOOL` error, whose message names the route back through `run_code` because the name IS declared to this model (an already-aborted caller signal keeps the cancellation contract: `ABORTED_BEFORE_DISPATCH`, with the visible tool's finalizer applied). The effective scope mode includes declarations inherited from an agent preset, so its wire schema and execution permissions remain aligned. A collapsed call terminates at `createExecution` — the first stage of `prepare` — BEFORE the extensible policy pipeline, so `tools/pre-execute` listeners, approval `ask`, and guards never observe a call that is deterministically denied; a human is never prompted to approve it. A nested sub-dispatch (`nested = true` — a `parent` token set, which only the `run_code` SDK binding sets in production code) may call any visible tool, so programs keep every binding the generated SDK declared.
 
 Four execution-path lookups — `executionMode`, `dispatchToolBody`, `postExecute`, `normalizeDispatchResult` — go through `resolveExecution`. `createExecution` applies the same collapse via the shared `collapses(name, nested)` predicate so it can distinguish a collapsed call from a genuinely unknown name before the policy pipeline. The public registry view (`get`) and SDK projection (`schemas`) keep their semantics: presentation, inspection, and binding enumeration still see the full visible set. The wire (`wireSchemas`) and the executor now agree. A collapsed call with non-JSON-serializable arguments reports the parameter `TypeError` (the invalid-args contract), not `UNKNOWN_TOOL` — the body still never runs and policy still does not.
 
@@ -41,5 +41,6 @@ No provider guarantees interception of unadvertised names; the reported session
 - `mode: 'code'` now enforces what it announces: a model-direct native call becomes `UNKNOWN_TOOL`, which the model can correct by routing through `run_code` (a pre-aborted call still resolves `ABORTED_BEFORE_DISPATCH`, per the cancellation contract).
 - `both` and `native` behavior is unchanged; SDK sub-dispatches are unchanged (the `parent` token is the discriminator).
 - A collapsed call is rejected at `prepare`, BEFORE the extensible policy pipeline: pre-execute listeners, approval `ask`, and guards never observe it. `executionMode` also fails closed (`exclusive`), so scheduling has no observable difference.
-- Under `code` mode, the imperative native-tool guidance sections (`tool:read`, `tool:write`, `tool:bash`, etc.) are filtered from the system prompt by a `system-prompt/assemble` listener so the model is never told to call a tool it cannot reach directly. The `tools:sdk` section (TypeScript bindings) remains, so programs can still use every tool through `run_code`.
+- Native-tool guidance sections (`tool:read`, `tool:write`, `tool:bash`, etc.) remain in the system prompt because they describe capabilities available through the generated SDK as well as native function calls, and several carry cross-tool routing policy (`read` over `bash cat`, `read` before `write` for the default fs-policy, `subagent` over `workflow`) that no single tool description can hold. The executor collapse, not prompt filtering, prevents model-direct native calls.
+- The prompt STATES the collapse, in the `tools:code-only` section ordered ahead of the 100-199 guidance band. Those sections name their tool without qualifying how it is reached, so a model that read only them emitted a native call, received `UNKNOWN_TOOL` for a tool the same prompt declared, and concluded the deployment was inconsistent rather than correcting itself. The denial carries the route for the same reason. `both` renders the rule empty: its native calls do execute, so stating it there would be false — which is why `both-mode-turn` no longer shares `code-mode-turn`'s expected prompt.
 - Any future composite transport that sets a `parent` token opts its sub-dispatches into the full table, matching the nested-call semantics the token already documents.

.agents/notes/implemented/bug-fix/2026-08-07-code-mode-executor-collapse.zh.md

@@ -12,7 +12,7 @@ Status: implemented
 
 ## 决策
 
-`ToolRegistry` 通过新增的私有 `resolveExecution(name, scope, nested)` 解析可执行定义，在拥有该决策的操作边界上应用模式塌缩。`code` 模式下，模型直呼（`nested = false`）只允许命名保留的 `run_code` 传输工具；任何原生名字都解析为 `undefined`，并以执行器既有的 `UNKNOWN_TOOL` 错误呈现（已中止的调用方 signal 保留取消契约：`ABORTED_BEFORE_DISPATCH`，并应用可见工具的 finalizer）。被塌缩的调用在 `createExecution`（`prepare` 的第一阶段）即终止——在可扩展策略流水线之前，因此 `tools/pre-execute` 监听器、approval `ask` 与 guard 永远不会观察到一个注定被拒绝的调用，人类也不会被提示去批准它。嵌套子调用（`nested = true`——即设置了 `parent` token，生产代码中只有 `run_code` SDK 绑定会设置）可以调用任意可见工具，因此程序保留生成 SDK 声明的全部绑定。
+`ToolRegistry` 通过新增的私有 `resolveExecution(name, scope, nested)` 解析可执行定义，在拥有该决策的操作边界上应用模式塌缩。当 `modeFor(scope)` 解析为 `code` 时，模型直呼（`nested = false`）只允许命名保留的 `run_code` 传输工具；任何原生名字都解析为 `undefined`，并以执行器既有的 `UNKNOWN_TOOL` 错误呈现，其消息会指出改走 `run_code` 的正确路径——因为这个名字对当前模型而言是**已声明过**的（已中止的调用方 signal 保留取消契约：`ABORTED_BEFORE_DISPATCH`，并应用可见工具的 finalizer）。有效的 scope 模式包括从 agent preset 继承的声明，因此其 wire schema 与执行权限保持一致。被塌缩的调用在 `createExecution`（`prepare` 的第一阶段）即终止——在可扩展策略流水线之前，因此 `tools/pre-execute` 监听器、approval `ask` 与 guard 永远不会观察到一个注定被拒绝的调用，人类也不会被提示去批准它。嵌套子调用（`nested = true`——即设置了 `parent` token，生产代码中只有 `run_code` SDK 绑定会设置）可以调用任意可见工具，因此程序保留生成 SDK 声明的全部绑定。
 
 执行链路的四处查表——`executionMode`、`dispatchToolBody`、`postExecute`、`normalizeDispatchResult`——改走 `resolveExecution`。`createExecution` 通过共享的 `collapses(name, nested)` 谓词应用同一塌缩，以便在策略流水线之前区分被塌缩的调用与真正未知的名字。公共注册表视图（`get`）与 SDK 投影（`schemas`）语义不变：展示、检查与绑定枚举仍看到完整可见集合。通告（`wireSchemas`）与执行器现在一致。带非 JSON 可序列化参数的塌缩调用报告参数 `TypeError`（invalid-args 契约），而非 `UNKNOWN_TOOL`——函数体仍不会运行，策略也不会执行。
 
@@ -41,5 +41,6 @@ guard 是可选的插件扩展；安全不变量不能依赖部署恰好组装
 - `mode: 'code'` 现在兑现其通告：模型直呼原生工具变为 `UNKNOWN_TOOL`，模型可以通过改走 `run_code` 自行纠正（已中止的调用仍按取消契约解析为 `ABORTED_BEFORE_DISPATCH`）。
 - `both` 与 `native` 行为不变；SDK 子调用不变（判别信号是 `parent` token）。
 - 被塌缩的调用在 `prepare` 阶段即被拒绝——在可扩展策略流水线之前：pre-execute 监听器、approval `ask` 与 guard 永远不会观察到它。`executionMode` 同样 fail-closed（`exclusive`），调度无可观察差异。
-- 在 `code` 模式下，命令式原生工具指引段（`tool:read`、`tool:write`、`tool:bash` 等）现在通过 `system-prompt/assemble` 监听器从系统提示词中过滤，模型不会再被告知去直接调用它无法触达的工具。`tools:sdk` 段（TypeScript 绑定）保留，程序仍可通过 `run_code` 使用所有工具。
+- 原生工具指引段（`tool:read`、`tool:write`、`tool:bash` 等）保留在系统提示词中，因为它们同时描述了通过生成 SDK 及原生函数调用可用的能力，其中若干段还承载着任何单个工具描述都装不下的跨工具路由策略（`read` 优先于 `bash cat`、默认 fs-policy 要求先 `read` 再 `write`、一两个委派用 `subagent` 而非 `workflow`）。防止模型直呼原生工具的是执行器塌缩，而非提示词过滤。
+- 提示词会**声明**这条塌缩，位于排在 100–199 指导段之前的 `tools:code-only` 段。那些段只写出工具名而不限定其可达方式，因此只读到它们的模型会发出原生调用，为一个同一份提示词刚刚声明过的工具收到 `UNKNOWN_TOOL`，进而判定部署不一致，而不是自行纠正。拒绝信息给出正确路径也是同一原因。`both` 下该规则渲染为空：它的原生调用确实会执行，在那里声明就是假话——这也是 `both-mode-turn` 不再与 `code-mode-turn` 共用期望提示词的原因。
 - 未来任何设置 `parent` token 的组合传输，其子调用自动走全表，与该 token 已有的嵌套调用语义一致。