Launch a background, read-only consultation with Codex or Claude to get an independent second opinion on a task, model, or workspace without interrupting your chat.
后台启动一次 Codex 或 Claude 的独立只读咨询。
| Name | Required | Description | Default |
|---|---|---|---|
| role | No | 独立审稿人 | |
| task | Yes | ||
| agent | Yes | ||
| model | No | ||
| effort | No | ||
| context | No | ||
| task_id | No | ||
| workspace | No | 当前原生聊天所打开工作区的绝对路径；每次调用按当前聊天动态选择，不绑定安装目录。 | |
| human_owner | No |
Changes observed during successful MCP inspections.
v0.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=false, so safety is covered. The description usefully adds that the run is background (async, presumably returning a task_id) and independent/isolated, but it omits lifecycle details (how to retrieve results, whether it blocks, authorization/owner semantics) despite a 9-param surface.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single tight, front-loaded sentence with no filler. However it is arguably under-specified rather than richly concise given the 9-param surface.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a 9-parameter, no-output-schema tool with near-zero parameter documentation, one sentence is far from sufficient. An agent cannot determine the role/effort/model semantics or how to retrieve the async result.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is only 11% (just the workspace path), yet the tool has 9 parameters including role, task, model, effort (9-value enum), context, task_id and human_owner that are entirely undocumented. The description mentions only 'agent' implicitly via Codex/Claude, leaving most parameter meaning to guesswork.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (start) plus resource (consultation) and scopes it as background and read-only, naming the two supported agents (Codex/Claude). This clearly distinguishes it from write/implementation siblings, though it doesn't explicitly name those siblings (e.g. start_codex_implementation, start_model_debate) to route the agent.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance on when to choose this over the many sibling start_* tools. The 'background' and 'read-only consultation' framing implies a use case, but no explicit when/when-not conditions or alternatives are given.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.