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

Annotations already declare readOnly, openWorld, idempotent, and non-destructive, but the description adds critical behavioral nuance: the distinction between could_not_verify (a failed check, not evidence) and unsupported (we looked and found no source). It also discloses that company-financial claims use a structured fast path with 'exact percent-delta math' while other claims go to a grounded pipeline with verbatim evidence and judgment. These details go well beyond annotations and cover failure modes, which is essential for an agent to trust the verdict.

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 long but each sentence adds value: trigger phrases, routing logic, return values, and error semantics. However, the purpose is not front-loaded—the first sentence is a list of paraphrases rather than a direct statement of what the tool does. The structure is somewhat dense, but not wasteful. It earns a 4 for being information-dense with minor structural inefficiency.

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 complex tool with two execution paths, a nuanced return value set, and specific error-handling rules, the description covers all necessary aspects: what it takes as input, how it routes, what it returns, what could_not_verify means, and what unsupported means. Even without an output schema, the description gives enough detail for an agent to invoke and interpret the result. It is highly complete.

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?

The input schema already provides full descriptions for both parameters (claim and tolerance_pct), so the baseline is 3. The description adds meaningful extra guidance: tolerance_pct overrides the implied tolerance and should be set to 1–2 for hallucination detection, and the default is capped at 5. This clarifies parameter intent beyond the schema, earning a 4.

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 performs natural-language claim verification against authoritative sources, with explicit trigger phrases ('fact check', 'verify the claim that…'). It distinguishes from siblings by positioning itself as a specialized claim-verification tool that replaces a multi-step pipeline, and it specifies two distinct verification paths (SEC EDGAR/XBRL for company financials, grounded pipeline for everything else). This goes beyond a simple 'validate' definition and leaves no ambiguity about the tool's purpose.

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 directly states when to use it: 'Use whenever the agent needs to check whether something a user said is factually correct.' It also provides specific conditions for routing (company-financial claims vs. any other factual claim) and notes that it replaces 4–6 sequential calls, implying a usage advantage over orchestrating separate lookups. No alternative tools are explicitly named, but the context is clear and effectively distinguishes when this tool is appropriate.

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