# Core Tool Architecture This folder owns Core's one local tool representation, process and Location registration, effective lookup, and settlement. ## Representations - `tool.ts` defines the opaque canonical `Tool.make({ description, input, output, execute, toModelOutput })` value. Shipped built-ins and plugin tools use the same type. - `tools.ts` exposes the registration-only `Tools.Service` view used by Location producers. - `registry.ts` stores only canonical Location registrations, derives definitions, invokes tools, and applies generic output bounding. Do not add a second executable entry type, registry-owned executor, authorization callback, output-path callback, or legacy normalization path. ## Construction Tool schemas and projection use `input` and `output` terminology. A tool value is opaque: its codecs, executor, definition derivation, and catalog permission declaration are private runtime details. Location-scoped built-in layers acquire `PermissionV2.Service` and every other required Location service while the layer is constructed. The executor captures those services. Permission sources are always constructed from the canonical invocation context: ```ts const source = { type: "tool" as const, messageID: context.assistantMessageID, callID: context.toolCallID, } ``` Leaves own resolution, permission, and side-effect ordering. Translate only expected typed errors into `ToolFailure`; do not use `catchCause`, because interruption and defects must survive. ## Registration Built-ins and plugin tools register through `Tools.Service.register({ [name]: tool })`. Registrations may provide a group, which flattens direct model names to `_`, and may be deferred from direct model exposure. Registrations are scoped: - The latest active same-placement registration wins. - Closing any registration removes only that registration and reveals the next active one. - An invocation captures the effective tool once settlement starts. `ToolRegistry.Service` is Location-scoped. Do not make the registry process-global or construct a separate application-tool service for each Location. ## Permissions The registry has no `PermissionV2.Service` dependency and performs no execution authorization. An internal built-in-only operation attaches a permission action solely to preserve whole-tool definition filtering; it is not part of public `Tool.make`. Most tools default to their registered name; `edit`, `write`, and `patch` declare the shared `edit` action. Definition filtering is catalog visibility, not execution authorization. A call still executes the captured leaf policy if it reaches settlement. ## Output Built-ins return complete validated domain output. `ToolRegistry.Materialization.settle` is the only execution and generic model-output bounding boundary and owns managed retention paths. Producer capture limits are separate. For example, Bash keeps `AppProcess.maxOutputBytes` and accurately reports stdout/stderr capture loss, but it does not run model-output truncation or return a managed `outputPath`. ## Current Gaps - MCP and future Session-scoped registrations still need an explicit canonical registration design. - The public Session result shape currently exposes managed `outputPaths`; full storage encapsulation requires a future opaque managed-output reference design.