evidence_collect

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

With no annotations provided, the description carries the full burden of behavioral disclosure, and it excels. It explicitly states that the tool runs read-only probes, returns the actual observed statuses, counts 'unknown' as never enforced, and flags self-attestation as a critical finding (provenance.self_attested). It also reveals the side effect of sending adversarial input to an endpoint when `authorized` is true. This is rich, honest, and highly transparent.

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?

The description is front-loaded with a one-sentence summary and then well-structured sections for output details, self-audit warning, and arguments. It is longer than average, but every sentence adds necessary information for a tool with complex side effects and seven parameters. It earns its length, though a few rephrasings could tighten it slightly without losing meaning.

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?

Given the tool's complexity (7 parameters, no annotations, high-risk behavioral probes) and the presence of an output schema, the description covers the essential context: what the tool returns, how to interpret results, the self-attestation pitfall, and the recommended follow-up (evidence_verify). The description is self-sufficient for an agent to decide whether and when to invoke this tool correctly.

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 0%, so the description must compensate for the lack of parameter documentation in the schema, and it does so completely. The 'Args' section explains the meaning and purpose of all seven parameters, including the role of `authorized` as a consent flag, the distinction between `pid` omission leading to self-inspection, and what `project_path` covers (MCP tool grants, manifests, etc.). This is exactly what the description should add over the bare schema.

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 clearly states the tool's purpose: 'Probe a running agent deployment and return an evidence bundle.' It goes beyond a generic verb+resource by explaining the output format (pass/fail/unknown per control) and explicitly differentiates itself from the sibling tool evidence_verify by directing users to feed the bundle to it for grading. This makes the tool's distinct role in the workflow unambiguous.

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 description gives explicit usage context: it tells users when to use this tool (to collect evidence) and, importantly, provides a clear caveat about when not to rely on it (self-attestation) with a named alternative ('run cinch collect out of band'). It also specifies when the `authorized` parameter must be true, which doubles as a prerequisite for endpoint probing. This level of when/when-not guidance is exemplary.

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