fix(tools): address Codex review of arg validation (PR 1)

- enum membership now checked uniformly for all SchemaTypes, mirroring the
  converter which emits `enum` regardless of type (was string-only)
- checkValue switch ends in assertNever per the closed-union convention
- sync the adding-a-tool cookbook to the validate-for-you behavior
- soften ADR 0011's property-test claim (RFC 001 not yet landed)
This commit is contained in:
Tianyi Cui
2026-06-13 23:11:48 +08:00
parent 36a30180b8
commit 11f85b4f88
4 changed files with 29 additions and 9 deletions
+1 -1
View File
@@ -15,6 +15,6 @@ The validator mirrors `schemaSpecToJsonSchema` semantics exactly — same struct
## Consequences
- The model gets actionable feedback on its own malformed calls instead of an opaque crash, closing the gap between `InferArgs`'s promise and runtime reality.
- The validator and `InferArgs` must stay in agreement; that drift risk is closed by a property test (RFC 001) generating args that satisfy `InferArgs` and asserting they pass `validateArgs`.
- The validator and `InferArgs` must stay in agreement; that drift risk is to be closed by a property test (RFC 001, not yet landed) generating args that satisfy `InferArgs` and asserting they pass `validateArgs`. Until then the agreement rests on the example tests and the shared converter structure.
- `ToolArgsError` is a plain `Error` with a `code` field for now; if a harness-wide error taxonomy lands it becomes a subclass without changing callers that read `.message`.
- Validation cost is negligible next to a model call.
+1 -1
View File
@@ -32,7 +32,7 @@ Registration is effect-based: disposing the plugin fiber unregisters the tool (w
## Rules of the execute() contract
- **Validate args at runtime.** `defineTool`'s `InferArgs` typing is compile-time only; at runtime `arguments` is whatever JSON the model emitted. Check every field; throw a descriptive Error for bad input.
- **Args are validated for you.** `defineTool` validates the model-generated `arguments` against the `SchemaSpec` before `execute` runs (type, required keys, enum membership, nested objects/arrays — ADR 0011), so inside `execute` the args already match `InferArgs`. You still hand-check value constraints the DSL can't express (non-empty strings, positive numbers, cross-field rules); throw a descriptive Error for those. Raw JSON-Schema tools registered directly (MCP) are NOT validated by the harness — they validate their own input.
- **Throwing means isError.** The registry catches anything `execute()` throws and returns `{isError: true}` to the model. Use that for infrastructure failures (bad input, spawn errors, aborts) — but REPORT domain failures in the result text instead (e.g. tool-bash returns `[exit code: 9]` with `isError: false`: the model decides what a failing command means).
- **Honor `exec.signal`.** Cancel in-flight work when it fires.
- **Use `exec.agent` for async notifications.** `agent.inject(content, {source: {kind: 'plugin', plugin: '<name>'}})` appends durable context the NEXT model request sees — it is not a wake-up (an idle agent stays idle). Guard against disposed agents (try/catch).
+15 -7
View File
@@ -20,6 +20,7 @@
*/
import type { ContentBlock } from '@deepseek-ai/dsh-llm'
import { assertNever } from '@deepseek-ai/dsh-llm'
import type { ToolDefinition, ToolExecution } from './index.ts'
// ---------------------------------------------------------------------------
@@ -202,16 +203,15 @@ function checkValue(prop: SchemaProp, value: unknown, path: string): string[] {
switch (prop.type) {
case 'string': {
if (typeof value !== 'string') return [`"${path}" must be a string`]
if (prop.enum && !prop.enum.includes(value)) {
return [`"${path}" must be one of ${JSON.stringify(prop.enum)}`]
}
return []
break
}
case 'number': {
return typeof value === 'number' ? [] : [`"${path}" must be a number`]
if (typeof value !== 'number') return [`"${path}" must be a number`]
break
}
case 'boolean': {
return typeof value === 'boolean' ? [] : [`"${path}" must be a boolean`]
if (typeof value !== 'boolean') return [`"${path}" must be a boolean`]
break
}
case 'object': {
if (!isPlainObject(value)) return [`"${path}" must be an object`]
@@ -225,8 +225,16 @@ function checkValue(prop: SchemaProp, value: unknown, path: string): string[] {
const items = prop.items
return value.flatMap((el, i) => checkValue(items, el, `${path}[${i}]`))
}
// No default: SchemaType is a closed union; every case is handled above.
default: return assertNever(prop.type, 'validateArgs')
}
// Enum membership, checked uniformly: the converter emits `enum` for any
// type ([prop.enum]), so the validator must too. `enum` is `string[]`, so a
// non-string value can never be a member — it falls out here, consistent
// with the schema the model was given.
if (prop.enum && !(prop.enum as unknown[]).includes(value)) {
return [`"${path}" must be one of ${JSON.stringify(prop.enum)}`]
}
return []
}
/** Collect violations for an object value against a {@link SchemaSpec}. */
+12
View File
@@ -623,6 +623,18 @@ describe('validateArgs (RFC 005 part 1)', () => {
expect(validateArgs(spec, { color: 'blue' })).toEqual(['"color" must be one of ["red","green"]'])
})
it('checks enum uniformly with the converter (enum on a non-string prop)', () => {
// The converter emits `enum` regardless of type; the validator must agree.
// `enum` is string[], so a number value can never be a member.
const spec = { n: { type: 'number', enum: ['1', '2'] } } as unknown as SchemaSpec
expect(validateArgs(spec, { n: 1 })).toEqual(['"n" must be one of ["1","2"]'])
})
it('rejects an unknown SchemaType at runtime (assertNever guard)', () => {
const spec = { x: { type: 'weird' } } as unknown as SchemaSpec
expect(() => validateArgs(spec, { x: 1 })).toThrow(/unreachable variant.*validateArgs/)
})
it('recurses into nested objects (and an object without properties only type-checks)', () => {
const spec = {
config: {