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

Beyond the annotations (readOnly, openWorld, idempotent, non-destructive), the description reveals significant behavioral details: the exact percent-delta math for financial claims, fallthrough routing, and the critical warning that could_not_verify means no verification happened and must not be presented as evidence. It also defines unsupported as a coverage gap, adding valuable nuance absent from annotations.

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 search-like trigger phrases, then moves from purpose to routing to output to caveats. Despite its length, every sentence adds unique information—no filler or redundancy—and the structure guides the reader from 'when' to 'how' to 'what to expect.'

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 (routing, multiple verdicts, error handling) and the absence of an output schema, the description compensates admirably. It covers expected return values, citation format, error semantics, and the distinction between could_not_verify and unsupported, making the tool fully understandable without an output schema.

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?

Although schema coverage is 100%, the description enriches both parameters: claim gets realistic examples, and tolerance_pct gains practical meaning (overrides claim wording, range 0.5–50, recommended 1–2 for hallucination detection, default cap 5). This is exactly the extra semantic value expected.

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 opens with natural-language trigger phrases and explicitly states 'natural-language claim verification against authoritative sources,' making the tool's purpose unmistakable. It details specific output verdicts and distinguishes itself from generic lookups by noting it replaces 4–6 sequential calls.

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?

It explicitly instructs to use the tool 'whenever the agent needs to check whether something a user said is factually correct,' and explains the routing behavior for company-financial claims versus any other factual claim. The caveat about could_not_verify vs unsupported further clarifies correct usage and interpretation.

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