mirror of
https://github.com/deepseek-ai/deepseek-harness
synced 2026-08-15 21:04:50 +00:00
Revert "test(goal): align master snapshots"
This reverts commit 028e63a9eee281a254f113f4721a67343e5521ca.
This commit is contained in:
Turtle
@@ -48,6 +48,7 @@ Ask the user a concise question when you need confirmation, a choice, or missing
|
||||
"description": "Questions to ask the user before continuing.",
|
||||
"items": {
|
||||
"type": "object",
|
||||
"additionalProperties": true,
|
||||
"properties": {
|
||||
"id": {
|
||||
"type": "string",
|
||||
@@ -66,6 +67,7 @@ Ask the user a concise question when you need confirmation, a choice, or missing
|
||||
"description": "Optional choices to show the user. If you recommend one, put it first and append \"(Recommended)\" to that label.",
|
||||
"items": {
|
||||
"type": "object",
|
||||
"additionalProperties": true,
|
||||
"properties": {
|
||||
"label": {
|
||||
"type": "string",
|
||||
@@ -229,7 +231,7 @@ Source: [`packages/cordis/tool-cordis/src/index.ts`](../packages/cordis/tool-cor
|
||||
|
||||
### `cordis_mount`
|
||||
|
||||
Mount a NEW cordis plugin into the live runtime that is running THIS agent (self-modification). `code` runs as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:"api" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:"events"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, async execute(args) { … } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'boolean'|'object'|'array', required?: true, description?, enum?, items?, properties? }; a JSON-Schema-style { type: 'object', properties, required: […] } wrapper and type 'integer' are also accepted and normalized. A tool's `execute` MUST return an ARRAY of content blocks, e.g. `return [{ type: 'text', text: someString }]` — never a bare string. Mounts can COMPOSE: one plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically on unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned on unmount) — cordis_inspect what:"api" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` VETOES the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.
|
||||
Mount a NEW cordis plugin into the live runtime that is running THIS agent (self-modification). `code` runs as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:"api" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:"events"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Mounts can COMPOSE: one plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically on unmount. Sandbox globals: `console` (tagged `[cordis:<id>]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned on unmount) — cordis_inspect what:"api" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` VETOES the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.
|
||||
|
||||
```json
|
||||
{
|
||||
@@ -628,7 +630,7 @@ Source: [`packages/goal/tool-goal/src/index.ts`](../packages/goal/tool-goal/src/
|
||||
|
||||
### `update_goal`
|
||||
|
||||
Update the exact current goal revision. edit, pause, and resume require a direct top-level human request. During an automatic continuation of the current goal, complete and blocked are also allowed. Fields not used by the selected action are ignored. blocked is rejected before the configured minimum round count; the model remains responsible for judging that the same condition persisted across those rounds and must explain it in blocked_reason.
|
||||
Update the exact current goal revision. edit, pause, and resume require a direct top-level human request. During an automatic continuation of the current goal, complete and blocked are also allowed. blocked is rejected before the configured minimum round count; the model remains responsible for judging that the same condition persisted across those rounds and must explain it in blocked_reason.
|
||||
|
||||
```json
|
||||
{
|
||||
@@ -655,15 +657,15 @@ Update the exact current goal revision. edit, pause, and resume require a direct
|
||||
},
|
||||
"objective": {
|
||||
"type": "string",
|
||||
"description": "Replacement objective for action edit; ignored for other actions."
|
||||
"description": "Replacement objective; valid only with action edit."
|
||||
},
|
||||
"max_goal_rounds": {
|
||||
"type": "number",
|
||||
"description": "Replacement round cap for action edit; ignored for other actions."
|
||||
"description": "Replacement cap; valid only with action edit."
|
||||
},
|
||||
"blocked_reason": {
|
||||
"type": "string",
|
||||
"description": "Concrete blocking condition for action blocked; required for blocked, ignored for other actions."
|
||||
"description": "Concrete blocking condition; required only with action blocked."
|
||||
}
|
||||
},
|
||||
"required": [
|
||||
@@ -896,6 +898,7 @@ Record and update a structured task list for the current work. Send the ENTIRE l
|
||||
"description": "The COMPLETE task list, replacing any previous list.",
|
||||
"items": {
|
||||
"type": "object",
|
||||
"additionalProperties": true,
|
||||
"properties": {
|
||||
"content": {
|
||||
"type": "string",
|
||||
@@ -937,7 +940,7 @@ Run a JavaScript workflow script that orchestrates subagents at scale. Use this
|
||||
The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result.
|
||||
|
||||
Script-body hooks:
|
||||
- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const — no oneOf/pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.
|
||||
- `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.
|
||||
- `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.
|
||||
- `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.
|
||||
- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.
|
||||
@@ -957,6 +960,7 @@ Constraints: concurrency and total-agent caps apply; no filesystem, network, tim
|
||||
"meta": {
|
||||
"type": "object",
|
||||
"description": "The workflow identity block (plain JSON — never code).",
|
||||
"additionalProperties": true,
|
||||
"properties": {
|
||||
"name": {
|
||||
"type": "string",
|
||||
@@ -975,6 +979,7 @@ Constraints: concurrency and total-agent caps apply; no filesystem, network, tim
|
||||
"description": "Optional phase declarations matched by phase() calls.",
|
||||
"items": {
|
||||
"type": "object",
|
||||
"additionalProperties": true,
|
||||
"properties": {
|
||||
"title": {
|
||||
"type": "string",
|
||||
@@ -1006,7 +1011,8 @@ Constraints: concurrency and total-agent caps apply; no filesystem, network, tim
|
||||
},
|
||||
"args": {
|
||||
"type": "object",
|
||||
"description": "Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]})."
|
||||
"description": "Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).",
|
||||
"additionalProperties": true
|
||||
}
|
||||
},
|
||||
"required": [
|
||||
|
||||
@@ -190,7 +190,7 @@ interface ToolArgsMap {
|
||||
status: "pending" | "in_progress" | "completed";
|
||||
} & Record<string, JsonValue>)[];
|
||||
} & Record<string, JsonValue>;
|
||||
/** Update the exact current goal revision. edit, pause, and resume require a direct top-level human request. During an automatic continuation of the current goal, complete and blocked are also allowed. Fields not used by the selected action are ignored. blocked is rejected before the configured minimum round count; the model remains responsible for judging that the same condition persisted across those rounds and must explain it in blocked_reason. */
|
||||
/** Update the exact current goal revision. edit, pause, and resume require a direct top-level human request. During an automatic continuation of the current goal, complete and blocked are also allowed. blocked is rejected before the configured minimum round count; the model remains responsible for judging that the same condition persisted across those rounds and must explain it in blocked_reason. */
|
||||
update_goal: {
|
||||
/** Exact id returned by get_goal. */
|
||||
goal_id: string;
|
||||
@@ -198,11 +198,11 @@ interface ToolArgsMap {
|
||||
revision: number;
|
||||
/** edit | pause | resume | complete | blocked */
|
||||
action: "edit" | "pause" | "resume" | "complete" | "blocked";
|
||||
/** Replacement objective for action edit; ignored for other actions. */
|
||||
/** Replacement objective; valid only with action edit. */
|
||||
objective?: string;
|
||||
/** Replacement round cap for action edit; ignored for other actions. */
|
||||
/** Replacement cap; valid only with action edit. */
|
||||
max_goal_rounds?: number;
|
||||
/** Concrete blocking condition for action blocked; required for blocked, ignored for other actions. */
|
||||
/** Concrete blocking condition; required only with action blocked. */
|
||||
blocked_reason?: string;
|
||||
} & Record<string, JsonValue>;
|
||||
/** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */
|
||||
|
||||
@@ -11,7 +11,7 @@
|
||||
{"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
|
||||
{"type":"assistant/message","seq":10,"time":0,"data":{"turn":1,"step":1,"content":[{"type":"tool-call","id":"call_spill","name":"bash","arguments":"{\"command\":\"node -e \\\"process.stdout.write('SPILL_START-' + 'x'.repeat(2000) + '-SPILL_END')\\\"\",\"description\":\"Print large deterministic output\"}"}],"provenance":{"provider":"deepseek","model":"deepseek-v4-flash"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[5,6,7,8,9],"surfaceOp":"append"}
|
||||
{"type":"tool/call","seq":11,"time":0,"data":{"turn":1,"step":1,"callId":"call_spill","name":"bash","arguments":"{\"command\":\"node -e \\\"process.stdout.write('SPILL_START-' + 'x'.repeat(2000) + '-SPILL_END')\\\"\",\"description\":\"Print large deterministic output\"}"}}
|
||||
{"type":"tool/result","seq":12,"time":0,"data":{"turn":1,"step":1,"callId":"call_spill","content":[{"type":"text","text":"SPILL_START-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx-SPILL_END\n\n(Omitted 1417 bytes. Full formatted result stored at: /tmp/dsh-acp-snap-ee77dff02/session-02b64799b331/1bc5704fddf2-bash.txt. Use read with offset/limit, or grep this path to search within it.)"}],"isError":false},"sourceEventSeqs":[11],"surfaceOp":"append"}
|
||||
{"type":"tool/result","seq":12,"time":0,"data":{"turn":1,"step":1,"callId":"call_spill","content":[{"type":"text","text":"SPILL_START-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx-SPILL_END\n\n(Omitted 1417 bytes. Full formatted result stored at: /tmp/dsh-acp-snap-ee77dff02/session-fbfcf2f560a0/1bddd2b64176-bash.txt. Use read with offset/limit, or grep this path to search within it.)"}],"isError":false},"sourceEventSeqs":[11],"surfaceOp":"append"}
|
||||
{"type":"step/end","seq":13,"time":0,"data":{"turn":1,"step":1}}
|
||||
{"type":"step/start","seq":14,"time":0,"data":{"turn":1,"step":2}}
|
||||
{"type":"assistant/chunk","seq":15,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
|
||||
|
||||
@@ -173,7 +173,7 @@ interface ToolArgsMap {
|
||||
status: "pending" | "in_progress" | "completed";
|
||||
} & Record<string, JsonValue>)[];
|
||||
} & Record<string, JsonValue>;
|
||||
/** Update the exact current goal revision. edit, pause, and resume require a direct top-level human request. During an automatic continuation of the current goal, complete and blocked are also allowed. Fields not used by the selected action are ignored. blocked is rejected before the configured minimum round count; the model remains responsible for judging that the same condition persisted across those rounds and must explain it in blocked_reason. */
|
||||
/** Update the exact current goal revision. edit, pause, and resume require a direct top-level human request. During an automatic continuation of the current goal, complete and blocked are also allowed. blocked is rejected before the configured minimum round count; the model remains responsible for judging that the same condition persisted across those rounds and must explain it in blocked_reason. */
|
||||
update_goal: {
|
||||
/** Exact id returned by get_goal. */
|
||||
goal_id: string;
|
||||
@@ -181,11 +181,11 @@ interface ToolArgsMap {
|
||||
revision: number;
|
||||
/** edit | pause | resume | complete | blocked */
|
||||
action: "edit" | "pause" | "resume" | "complete" | "blocked";
|
||||
/** Replacement objective for action edit; ignored for other actions. */
|
||||
/** Replacement objective; valid only with action edit. */
|
||||
objective?: string;
|
||||
/** Replacement round cap for action edit; ignored for other actions. */
|
||||
/** Replacement cap; valid only with action edit. */
|
||||
max_goal_rounds?: number;
|
||||
/** Concrete blocking condition for action blocked; required for blocked, ignored for other actions. */
|
||||
/** Concrete blocking condition; required only with action blocked. */
|
||||
blocked_reason?: string;
|
||||
} & Record<string, JsonValue>;
|
||||
/** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */
|
||||
|
||||
@@ -173,7 +173,7 @@ interface ToolArgsMap {
|
||||
status: "pending" | "in_progress" | "completed";
|
||||
} & Record<string, JsonValue>)[];
|
||||
} & Record<string, JsonValue>;
|
||||
/** Update the exact current goal revision. edit, pause, and resume require a direct top-level human request. During an automatic continuation of the current goal, complete and blocked are also allowed. Fields not used by the selected action are ignored. blocked is rejected before the configured minimum round count; the model remains responsible for judging that the same condition persisted across those rounds and must explain it in blocked_reason. */
|
||||
/** Update the exact current goal revision. edit, pause, and resume require a direct top-level human request. During an automatic continuation of the current goal, complete and blocked are also allowed. blocked is rejected before the configured minimum round count; the model remains responsible for judging that the same condition persisted across those rounds and must explain it in blocked_reason. */
|
||||
update_goal: {
|
||||
/** Exact id returned by get_goal. */
|
||||
goal_id: string;
|
||||
@@ -181,11 +181,11 @@ interface ToolArgsMap {
|
||||
revision: number;
|
||||
/** edit | pause | resume | complete | blocked */
|
||||
action: "edit" | "pause" | "resume" | "complete" | "blocked";
|
||||
/** Replacement objective for action edit; ignored for other actions. */
|
||||
/** Replacement objective; valid only with action edit. */
|
||||
objective?: string;
|
||||
/** Replacement round cap for action edit; ignored for other actions. */
|
||||
/** Replacement cap; valid only with action edit. */
|
||||
max_goal_rounds?: number;
|
||||
/** Concrete blocking condition for action blocked; required for blocked, ignored for other actions. */
|
||||
/** Concrete blocking condition; required only with action blocked. */
|
||||
blocked_reason?: string;
|
||||
} & Record<string, JsonValue>;
|
||||
/** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */
|
||||
|
||||
@@ -173,7 +173,7 @@ interface ToolArgsMap {
|
||||
status: "pending" | "in_progress" | "completed";
|
||||
} & Record<string, JsonValue>)[];
|
||||
} & Record<string, JsonValue>;
|
||||
/** Update the exact current goal revision. edit, pause, and resume require a direct top-level human request. During an automatic continuation of the current goal, complete and blocked are also allowed. Fields not used by the selected action are ignored. blocked is rejected before the configured minimum round count; the model remains responsible for judging that the same condition persisted across those rounds and must explain it in blocked_reason. */
|
||||
/** Update the exact current goal revision. edit, pause, and resume require a direct top-level human request. During an automatic continuation of the current goal, complete and blocked are also allowed. blocked is rejected before the configured minimum round count; the model remains responsible for judging that the same condition persisted across those rounds and must explain it in blocked_reason. */
|
||||
update_goal: {
|
||||
/** Exact id returned by get_goal. */
|
||||
goal_id: string;
|
||||
@@ -181,11 +181,11 @@ interface ToolArgsMap {
|
||||
revision: number;
|
||||
/** edit | pause | resume | complete | blocked */
|
||||
action: "edit" | "pause" | "resume" | "complete" | "blocked";
|
||||
/** Replacement objective for action edit; ignored for other actions. */
|
||||
/** Replacement objective; valid only with action edit. */
|
||||
objective?: string;
|
||||
/** Replacement round cap for action edit; ignored for other actions. */
|
||||
/** Replacement cap; valid only with action edit. */
|
||||
max_goal_rounds?: number;
|
||||
/** Concrete blocking condition for action blocked; required for blocked, ignored for other actions. */
|
||||
/** Concrete blocking condition; required only with action blocked. */
|
||||
blocked_reason?: string;
|
||||
} & Record<string, JsonValue>;
|
||||
/** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return <value>` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise<any>` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise<any[]>` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise<any[]>` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */
|
||||
|
||||
@@ -132,8 +132,8 @@
|
||||
{"type":"assistant/chunk","seq":130,"time":1783962245385,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
|
||||
{"type":"assistant/message","seq":131,"time":1783962245385,"data":{"turn":1,"step":1,"content":[{"type":"reasoning","text":"The user wants me to run a command with sandbox_permissions set to danger-full-access, no prior run needed, justified as instructed."},{"type":"tool-call","id":"call_00_d0sAHpJ9mYOJi0z7KNy30441","name":"bash","arguments":"{\"command\": \"printf 'escalated\\\\n' > /tmp/dsh-escalated.txt && cat /tmp/dsh-escalated.txt && rm /tmp/dsh-escalated.txt\", \"description\": \"Write file outside workspace and verify\", \"sandbox_permissions\": \"danger-full-access\", \"justification\": \"the user asked to write a file outside the workspace\"}"}],"provenance":{"provider":"deepseek","model":"deepseek-v4-flash"},"usage":{"inputTokens":1501,"outputTokens":174,"cacheReadTokens":0,"reasoningTokens":28}},"sourceEventSeqs":[8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130],"surfaceOp":"append"}
|
||||
{"type":"tool/call","seq":132,"time":1783962245385,"data":{"turn":1,"step":1,"callId":"call_00_d0sAHpJ9mYOJi0z7KNy30441","name":"bash","arguments":"{\"command\": \"printf 'escalated\\\\n' > /tmp/dsh-escalated.txt && cat /tmp/dsh-escalated.txt && rm /tmp/dsh-escalated.txt\", \"description\": \"Write file outside workspace and verify\", \"sandbox_permissions\": \"danger-full-access\", \"justification\": \"the user asked to write a file outside the workspace\"}"}}
|
||||
{"type":"approval/asked","seq":133,"time":1783962245386,"data":{"id":"bb75e939-221a-48b0-ad67-3da110ecbd31","toolName":"bash","callId":"call_00_d0sAHpJ9mYOJi0z7KNy30441","reason":"escalate sandbox to danger-full-access: the user asked to write a file outside the workspace"}}
|
||||
{"type":"approval/decided","seq":134,"time":1783962245387,"data":{"id":"bb75e939-221a-48b0-ad67-3da110ecbd31","outcome":"allowed-once"}}
|
||||
{"type":"approval/asked","seq":133,"time":1783962245386,"data":{"id":"21634890-d003-489f-b1b5-d38e187561c5","toolName":"bash","callId":"call_00_d0sAHpJ9mYOJi0z7KNy30441","reason":"escalate sandbox to danger-full-access: the user asked to write a file outside the workspace"}}
|
||||
{"type":"approval/decided","seq":134,"time":1783962245387,"data":{"id":"21634890-d003-489f-b1b5-d38e187561c5","outcome":"allowed-once"}}
|
||||
{"type":"tool/result","seq":135,"time":1783962245399,"data":{"turn":1,"step":1,"callId":"call_00_d0sAHpJ9mYOJi0z7KNy30441","content":[{"type":"text","text":"escalated\n"}],"isError":false},"sourceEventSeqs":[132],"surfaceOp":"append"}
|
||||
{"type":"step/end","seq":136,"time":1783962245400,"data":{"turn":1,"step":1}}
|
||||
{"type":"step/start","seq":137,"time":1783962245400,"data":{"turn":1,"step":2}}
|
||||
|
||||
@@ -156,8 +156,8 @@
|
||||
{"type":"assistant/chunk","seq":154,"time":1783962246274,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
|
||||
{"type":"assistant/message","seq":155,"time":1783962246274,"data":{"turn":1,"step":1,"content":[{"type":"reasoning","text":"The user wants me to run a specific command with `sandbox_permissions` set to `danger-full-access` and a specific justification. They explicitly said NOT to run it without sandbox_permissions first. Let me do exactly that."},{"type":"tool-call","id":"call_00_WB1vnPomi8yr6MlcFKTj7912","name":"bash","arguments":"{\"command\": \"printf 'escalated\\\\n' > /tmp/dsh-escalated.txt && cat /tmp/dsh-escalated.txt && rm /tmp/dsh-escalated.txt\", \"description\": \"Write to /tmp and verify, then clean up\", \"sandbox_permissions\": \"danger-full-access\", \"justification\": \"the user asked to write a file outside the workspace\"}"}],"provenance":{"provider":"deepseek","model":"deepseek-v4-flash"},"usage":{"inputTokens":1509,"outputTokens":198,"cacheReadTokens":0,"reasoningTokens":48}},"sourceEventSeqs":[8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154],"surfaceOp":"append"}
|
||||
{"type":"tool/call","seq":156,"time":1783962246274,"data":{"turn":1,"step":1,"callId":"call_00_WB1vnPomi8yr6MlcFKTj7912","name":"bash","arguments":"{\"command\": \"printf 'escalated\\\\n' > /tmp/dsh-escalated.txt && cat /tmp/dsh-escalated.txt && rm /tmp/dsh-escalated.txt\", \"description\": \"Write to /tmp and verify, then clean up\", \"sandbox_permissions\": \"danger-full-access\", \"justification\": \"the user asked to write a file outside the workspace\"}"}}
|
||||
{"type":"approval/asked","seq":157,"time":1783962246275,"data":{"id":"9a997191-81a0-4076-914b-255ebe3e6882","toolName":"bash","callId":"call_00_WB1vnPomi8yr6MlcFKTj7912","reason":"escalate sandbox to danger-full-access: the user asked to write a file outside the workspace"}}
|
||||
{"type":"approval/decided","seq":158,"time":1783962246275,"data":{"id":"9a997191-81a0-4076-914b-255ebe3e6882","outcome":"rejected"}}
|
||||
{"type":"approval/asked","seq":157,"time":1783962246275,"data":{"id":"4ffb95e3-8f94-41db-bfd3-2f56e6631ef8","toolName":"bash","callId":"call_00_WB1vnPomi8yr6MlcFKTj7912","reason":"escalate sandbox to danger-full-access: the user asked to write a file outside the workspace"}}
|
||||
{"type":"approval/decided","seq":158,"time":1783962246275,"data":{"id":"4ffb95e3-8f94-41db-bfd3-2f56e6631ef8","outcome":"rejected"}}
|
||||
{"type":"tool/result","seq":159,"time":1783962246275,"data":{"turn":1,"step":1,"callId":"call_00_WB1vnPomi8yr6MlcFKTj7912","content":[{"type":"text","text":"Error: the user rejected escalating this command to \"danger-full-access\""}],"isError":true},"sourceEventSeqs":[156],"surfaceOp":"append"}
|
||||
{"type":"step/end","seq":160,"time":1783962246276,"data":{"turn":1,"step":1}}
|
||||
{"type":"step/start","seq":161,"time":1783962246276,"data":{"turn":1,"step":2}}
|
||||
|
||||
@@ -90,9 +90,9 @@
|
||||
{"type":"assistant/chunk","seq":88,"time":1784045703776,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
|
||||
{"type":"assistant/message","seq":89,"time":1784045703780,"data":{"turn":1,"step":1,"content":[{"type":"reasoning","text":"The user wants me to create a file using the write tool with sandbox_permissions. Let me do that."},{"type":"tool-call","id":"call_00_Fnymmavpr4klMDy4Fdej3227","name":"write","arguments":"{\"file_path\": \"escalated.md\", \"content\": \"escalated\", \"sandbox_permissions\": \"danger-full-access\", \"justification\": \"the user asked to escalate this write\"}"}],"provenance":{"provider":"deepseek","model":"deepseek-v4-flash"},"usage":{"inputTokens":3871,"outputTokens":132,"cacheReadTokens":0,"reasoningTokens":23}},"sourceEventSeqs":[8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88],"surfaceOp":"append"}
|
||||
{"type":"tool/call","seq":90,"time":1784045703780,"data":{"turn":1,"step":1,"callId":"call_00_Fnymmavpr4klMDy4Fdej3227","name":"write","arguments":"{\"file_path\": \"escalated.md\", \"content\": \"escalated\", \"sandbox_permissions\": \"danger-full-access\", \"justification\": \"the user asked to escalate this write\"}"}}
|
||||
{"type":"approval/asked","seq":91,"time":1784045703782,"data":{"id":"f6a28337-87da-4555-bc47-2b5354300cc4","toolName":"write","callId":"call_00_Fnymmavpr4klMDy4Fdej3227","reason":"escalate sandbox to danger-full-access: the user asked to escalate this write"}}
|
||||
{"type":"approval/decided","seq":92,"time":1784045703786,"data":{"id":"f6a28337-87da-4555-bc47-2b5354300cc4","outcome":"allowed-once"}}
|
||||
{"type":"tool/result","seq":93,"time":1784045703798,"data":{"turn":1,"step":1,"callId":"call_00_Fnymmavpr4klMDy4Fdej3227","content":[{"type":"text","text":"<path>/private/var/folders/2g/b32ct0qn1d728l_v6tdkjytr0000gn/T/acp-snap-cwd-vmEGzd/escalated.md</path>\n<type>file</type>\n<content>\nCreated file\n</content>"}],"isError":false,"meta":{"diffs":[]}},"sourceEventSeqs":[90],"surfaceOp":"append"}
|
||||
{"type":"approval/asked","seq":91,"time":1784045703782,"data":{"id":"d7250dff-fc08-4163-9b51-f43f08d6f36f","toolName":"write","callId":"call_00_Fnymmavpr4klMDy4Fdej3227","reason":"escalate sandbox to danger-full-access: the user asked to escalate this write"}}
|
||||
{"type":"approval/decided","seq":92,"time":1784045703786,"data":{"id":"d7250dff-fc08-4163-9b51-f43f08d6f36f","outcome":"allowed-once"}}
|
||||
{"type":"tool/result","seq":93,"time":1784045703798,"data":{"turn":1,"step":1,"callId":"call_00_Fnymmavpr4klMDy4Fdej3227","content":[{"type":"text","text":"<path>/var/folders/2g/b32ct0qn1d728l_v6tdkjytr0000gn/T/acp-snap-cwd-vmEGzd/escalated.md</path>\n<type>file</type>\n<content>\nCreated file\n</content>"}],"isError":false,"meta":{"diffs":[]}},"sourceEventSeqs":[90],"surfaceOp":"append"}
|
||||
{"type":"step/end","seq":94,"time":1784045703798,"data":{"turn":1,"step":1}}
|
||||
{"type":"step/start","seq":95,"time":1784045703799,"data":{"turn":1,"step":2}}
|
||||
{"type":"assistant/chunk","seq":96,"time":1784045704512,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"reasoning"}}}
|
||||
|
||||
@@ -63,7 +63,7 @@
|
||||
{"type":"assistant/chunk","seq":61,"time":1783352079886,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
|
||||
{"type":"assistant/message","seq":62,"time":1783352079888,"data":{"turn":1,"step":1,"content":[{"type":"reasoning","text":"The user wants me to create a file named notes.txt with the content \"hello world\" using the write tool, then reply with \"DONE\"."},{"type":"tool-call","id":"call_00_APMUCJJm9lrTSlVbg6dB0185","name":"write","arguments":"{\"file_path\": \"notes.txt\", \"content\": \"hello world\"}"}],"provenance":{"provider":"deepseek","model":"deepseek-v4-flash"},"usage":{"inputTokens":2891,"outputTokens":92,"cacheReadTokens":0,"reasoningTokens":30}},"sourceEventSeqs":[5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61],"surfaceOp":"append"}
|
||||
{"type":"tool/call","seq":63,"time":1783352079888,"data":{"turn":1,"step":1,"callId":"call_00_APMUCJJm9lrTSlVbg6dB0185","name":"write","arguments":"{\"file_path\": \"notes.txt\", \"content\": \"hello world\"}"}}
|
||||
{"type":"tool/result","seq":64,"time":1783352079897,"data":{"turn":1,"step":1,"callId":"call_00_APMUCJJm9lrTSlVbg6dB0185","content":[{"type":"text","text":"<path>/private/tmp/acp-snap-cwd-sNvn5N/notes.txt</path>\n<type>file</type>\n<content>\nCreated file\n</content>"}],"isError":false,"meta":{"diffs":[]}},"sourceEventSeqs":[63],"surfaceOp":"append"}
|
||||
{"type":"tool/result","seq":64,"time":1783352079897,"data":{"turn":1,"step":1,"callId":"call_00_APMUCJJm9lrTSlVbg6dB0185","content":[{"type":"text","text":"<path>/tmp/acp-snap-cwd-sNvn5N/notes.txt</path>\n<type>file</type>\n<content>\nCreated file\n</content>"}],"isError":false,"meta":{"diffs":[]}},"sourceEventSeqs":[63],"surfaceOp":"append"}
|
||||
{"type":"step/end","seq":65,"time":1783352079898,"data":{"turn":1,"step":1}}
|
||||
{"type":"step/start","seq":66,"time":1783352079899,"data":{"turn":1,"step":2}}
|
||||
{"type":"assistant/chunk","seq":67,"time":1783352080825,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"reasoning"}}}
|
||||
|
||||
@@ -56,8 +56,8 @@
|
||||
{"type":"tool/call","seq":54,"time":1783352172557,"data":{"turn":1,"step":1,"callId":"call_00_6k0oGSliVHxGSgqBmMEO4311","name":"bash","arguments":"{\"command\": \"echo HELLO\", \"description\": \"Echo HELLO\"}"}}
|
||||
{"type":"hook/invoked","seq":55,"time":1783352172558,"data":{"turn":1,"point":"PreToolUse","dialect":"claude","handlerId":"claude:PreToolUse:1","matcher":"bash"}}
|
||||
{"type":"hook/result","seq":56,"time":1783352172573,"data":{"turn":1,"point":"PreToolUse","handlerId":"claude:PreToolUse:1","decision":"ask","exitCode":0,"durationMs":14.113374999999905}}
|
||||
{"type":"approval/asked","seq":57,"time":1783962235813,"data":{"id":"c7877002-3d2e-432e-9624-0b9ef3624df3","toolName":"bash","callId":"call_00_6k0oGSliVHxGSgqBmMEO4311","reason":"bash requires manual approval in this session"}}
|
||||
{"type":"approval/decided","seq":58,"time":1783962235813,"data":{"id":"c7877002-3d2e-432e-9624-0b9ef3624df3","outcome":"rejected"}}
|
||||
{"type":"approval/asked","seq":57,"time":1783962235813,"data":{"id":"e629ff8c-60d2-49b1-9603-333b16ef3306","toolName":"bash","callId":"call_00_6k0oGSliVHxGSgqBmMEO4311","reason":"bash requires manual approval in this session"}}
|
||||
{"type":"approval/decided","seq":58,"time":1783962235813,"data":{"id":"e629ff8c-60d2-49b1-9603-333b16ef3306","outcome":"rejected"}}
|
||||
{"type":"tool/result","seq":59,"time":1783962235814,"data":{"turn":1,"step":1,"callId":"call_00_6k0oGSliVHxGSgqBmMEO4311","content":[{"type":"text","text":"Error: the user rejected tool \"bash\""}],"isError":true},"sourceEventSeqs":[54],"surfaceOp":"append"}
|
||||
{"type":"step/end","seq":60,"time":1783962235814,"data":{"turn":1,"step":1}}
|
||||
{"type":"step/start","seq":61,"time":1783962235814,"data":{"turn":1,"step":2}}
|
||||
|
||||
@@ -324,7 +324,7 @@
|
||||
{"type":"assistant/chunk","seq":322,"time":1784525379646,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
|
||||
{"type":"assistant/message","seq":323,"time":1784525379647,"data":{"turn":1,"step":3,"content":[{"type":"reasoning","text":"The plan is approved. Now I need to apply exactly that one edit: change line 2 of notes.txt to say \"hello world\". I'll use the edit tool to replace the current line 2 content."},{"type":"tool-call","id":"call_00_eYuZ6fAuzqr8vtFd3oRg2452","name":"edit","arguments":"{\"file_path\": \"notes.txt\", \"old_string\": \"- the greeting message still says \\\"helo wrld\\\"\", \"new_string\": \"hello world\"}"}],"provenance":{"provider":"deepseek","model":"deepseek-v4-flash"},"usage":{"inputTokens":3269,"outputTokens":134,"cacheReadTokens":0,"reasoningTokens":43}},"sourceEventSeqs":[232,233,234,235,236,237,238,239,240,241,242,243,244,245,246,247,248,249,250,251,252,253,254,255,256,257,258,259,260,261,262,263,264,265,266,267,268,269,270,271,272,273,274,275,276,277,278,279,280,281,282,283,284,285,286,287,288,289,290,291,292,293,294,295,296,297,298,299,300,301,302,303,304,305,306,307,308,309,310,311,312,313,314,315,316,317,318,319,320,321,322],"surfaceOp":"append"}
|
||||
{"type":"tool/call","seq":324,"time":1784525379647,"data":{"turn":1,"step":3,"callId":"call_00_eYuZ6fAuzqr8vtFd3oRg2452","name":"edit","arguments":"{\"file_path\": \"notes.txt\", \"old_string\": \"- the greeting message still says \\\"helo wrld\\\"\", \"new_string\": \"hello world\"}"}}
|
||||
{"type":"tool/result","seq":325,"time":1784525379652,"data":{"turn":1,"step":3,"callId":"call_00_eYuZ6fAuzqr8vtFd3oRg2452","content":[{"type":"text","text":"Error: edit requires reading \"/private/var/folders/2g/b32ct0qn1d728l_v6tdkjytr0000gn/T/acp-snap-cwd-tia05W/notes.txt\" first"}],"isError":true,"error":{"name":"FsError","code":"FS_NOT_OBSERVED"}},"sourceEventSeqs":[324],"surfaceOp":"append"}
|
||||
{"type":"tool/result","seq":325,"time":1784525379652,"data":{"turn":1,"step":3,"callId":"call_00_eYuZ6fAuzqr8vtFd3oRg2452","content":[{"type":"text","text":"Error: edit requires reading \"/var/folders/2g/b32ct0qn1d728l_v6tdkjytr0000gn/T/acp-snap-cwd-tia05W/notes.txt\" first"}],"isError":true,"error":{"name":"FsError","code":"FS_NOT_OBSERVED"}},"sourceEventSeqs":[324],"surfaceOp":"append"}
|
||||
{"type":"step/end","seq":326,"time":1784525379652,"data":{"turn":1,"step":3}}
|
||||
{"type":"step/start","seq":327,"time":1784525379655,"data":{"turn":1,"step":4}}
|
||||
{"type":"assistant/chunk","seq":328,"time":1784525380205,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"reasoning"}}}
|
||||
@@ -428,7 +428,7 @@
|
||||
{"type":"assistant/chunk","seq":426,"time":1784525381745,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
|
||||
{"type":"assistant/message","seq":427,"time":1784525381746,"data":{"turn":1,"step":5,"content":[{"type":"reasoning","text":"Now I can apply the edit."},{"type":"tool-call","id":"call_00_7H6VFdLjBqzzW9M5Z4OR8016","name":"edit","arguments":"{\"file_path\": \"notes.txt\", \"old_string\": \"- the greeting message still says \\\"helo wrld\\\"\", \"new_string\": \"hello world\"}"}],"provenance":{"provider":"deepseek","model":"deepseek-v4-flash"},"usage":{"inputTokens":169,"outputTokens":98,"cacheReadTokens":3456,"reasoningTokens":7}},"sourceEventSeqs":[372,373,374,375,376,377,378,379,380,381,382,383,384,385,386,387,388,389,390,391,392,393,394,395,396,397,398,399,400,401,402,403,404,405,406,407,408,409,410,411,412,413,414,415,416,417,418,419,420,421,422,423,424,425,426],"surfaceOp":"append"}
|
||||
{"type":"tool/call","seq":428,"time":1784525381746,"data":{"turn":1,"step":5,"callId":"call_00_7H6VFdLjBqzzW9M5Z4OR8016","name":"edit","arguments":"{\"file_path\": \"notes.txt\", \"old_string\": \"- the greeting message still says \\\"helo wrld\\\"\", \"new_string\": \"hello world\"}"}}
|
||||
{"type":"tool/result","seq":429,"time":1784525381766,"data":{"turn":1,"step":5,"callId":"call_00_7H6VFdLjBqzzW9M5Z4OR8016","content":[{"type":"text","text":"The file /private/var/folders/2g/b32ct0qn1d728l_v6tdkjytr0000gn/T/acp-snap-cwd-tia05W/notes.txt has been updated successfully."}],"isError":false,"meta":{"diffs":[{"path":"notes.txt","oldText":"project notes\n- the greeting message still says \"helo wrld\"","newText":"project notes\nhello world"}]}},"sourceEventSeqs":[428],"surfaceOp":"append"}
|
||||
{"type":"tool/result","seq":429,"time":1784525381766,"data":{"turn":1,"step":5,"callId":"call_00_7H6VFdLjBqzzW9M5Z4OR8016","content":[{"type":"text","text":"The file /var/folders/2g/b32ct0qn1d728l_v6tdkjytr0000gn/T/acp-snap-cwd-tia05W/notes.txt has been updated successfully."}],"isError":false,"meta":{"diffs":[{"path":"notes.txt","oldText":"project notes\n- the greeting message still says \"helo wrld\"","newText":"project notes\nhello world"}]}},"sourceEventSeqs":[428],"surfaceOp":"append"}
|
||||
{"type":"step/end","seq":430,"time":1784525381766,"data":{"turn":1,"step":5}}
|
||||
{"type":"step/start","seq":431,"time":1784525381768,"data":{"turn":1,"step":6}}
|
||||
{"type":"assistant/chunk","seq":432,"time":1784525382334,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"reasoning"}}}
|
||||
|
||||
@@ -409,18 +409,10 @@ describe('goal tool state transitions', () => {
|
||||
}
|
||||
|
||||
const paused = await execute(ctx, 'update_goal', {
|
||||
goal_id: goal.id,
|
||||
revision: goal.revision,
|
||||
action: 'pause',
|
||||
objective: 'ignored replacement',
|
||||
max_goal_rounds: 1,
|
||||
blocked_reason: 'ignored blocker',
|
||||
goal_id: goal.id, revision: goal.revision, action: 'pause', ...strictArgs,
|
||||
}, root.agent)
|
||||
goal = ctx.goals.get(root.agent)!
|
||||
expect(resultGoal(paused)).toMatchObject({
|
||||
phase: 'paused', objective: 'valid', maxGoalRounds: goal.maxGoalRounds,
|
||||
})
|
||||
expect(goal.blockedReason).toBeUndefined()
|
||||
expect(resultGoal(paused)).toMatchObject({ phase: 'paused', objective: 'valid' })
|
||||
const resumed = await execute(ctx, 'update_goal', {
|
||||
goal_id: goal.id, revision: goal.revision, action: 'resume', ...strictArgs,
|
||||
}, root.agent)
|
||||
|
||||
Reference in New Issue
Block a user