blueprint_audit

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=false, destructiveHint=false, and openWorldHint=true. The description adds real value beyond that: it discloses cloud inference (data leaves the local machine, consistent with openWorldHint), the async contract (returns a task ID and requires polling status), and that output is a gap report plus PRD. It omits cost/rate implications for max_calls/max_tokens, but the async and cloud traits are useful non-obvious context.

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?

Three short sentences, front-loaded with the action and followed by the limitation rationale and the return/polling contract. Efficient, though the middle value-claim sentence is slightly promotional rather than strictly operational.

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?

There is no output schema, and the description does tell the agent it gets a task ID and must poll status, which is the key return behavior. But for a 4-parameter, open-world, non-read-only tool with only 25% param coverage, it leaves the parameter semantics and side effects (where artifacts are written, cost of cloud inference) underspecified.

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 25%: project_path has a description, but prd_path, max_calls, and max_tokens are undocumented in both schema and description. The description implies a project is scanned and a PRD is produced but never explains these parameters or their bounds, so it fails to compensate for the coverage gap.

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?

The description states a specific verb+resource: launching a real reverse-engineering blueprint that scans a local project and performs cloud inference to produce a gap report and a target PRD. This clearly separates it from a passive check, but it never names or contrasts the closest siblings (e.g., blueprint_plan or blueprint_health), leaving the agent to infer the distinction.

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?

The line '静态扫描不能证明生产可用' implies the audit is for cases where static scanning is insufficient, and '需轮询 status' hints at the async workflow via blueprint_status. However, there is no explicit when-to-use/when-not or direct routing to alternatives, so usage is only implied.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.