Verify a claim

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

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint, covering the safety profile. The description adds the behavioral detail that the tool categorizes evidence and proposes an experiment, which is beyond annotations. However, it does not disclose what happens if no evidence is supplied, or whether the proposed experiment is actionable output or just advice. With annotations handling safety, a 3 is appropriate – the description adds some process context but not deep behavioral nuance.

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 a single, efficient sentence that front-loads the core purpose. No wasted words, and it conveys the two main phases (categorization and proposal). It is slightly abstract but structurally sound. Loses a point for not being more concrete about the tool's operation.

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 tool with 6 parameters, a nested evidence array, and no output schema, the description is too high-level. It does not explain how an agent should structure input, what kind of 'decisive experiment' is produced, or what the return value looks like. The schema covers parameters, but the overall workflow and expected output remain ambiguous. An agent would struggle to know what to pass beyond the required claim and how to interpret the 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 100%, with each parameter having its own description (e.g., 'claim' described as 'The technical claim to examine'). The tool description adds no parameter-specific details, leaving the schema to carry all parameter semantics. Baseline 3 is correct since the schema fully documents inputs and the description does not contradict or supplement it.

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 action: separating claim-related information into categories and proposing a decisive experiment. The verb 'Separate' combined with 'propose' gives a precise enough purpose. It does not explicitly contrast with sibling tools like ackrite or reality_check, but the title 'Verify a claim' reinforces the intent. Scores 4 rather than 5 because the description could be more explicit about the tool's primary function being analysis of a claim rather than external verification.

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 provides no guidance on when to use this tool versus its siblings (ackrite, audit, prove_it, reality_check). It does not mention prerequisites, alternative tools, or scenarios where verification should be attempted via other means. The only implicit usage context is the word 'verify', but that is weak and does not help an agent choose among similar tools.

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