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

Annotations already cover the safety profile (readOnlyHint, idempotentHint, openWorldHint, non-destructive), and the description adds substantial behavioral context on top: the exact success payload ({answer, evidence, confidence, source, fetched_at, refusal_reason:null}), the full enumerated refusal_reason values, the constraint that answers are extracted ONLY from tool results, and the cost trade-off. The description also states a key behavioral trait — explicit refusal instead of guessing — which is critical for an agent deciding whether to trust the output. No contradiction with 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 the core purpose and the return contract appears early, before usage guidance. It is longer than average, but every sentence earns its place: there is no output schema, so the success/refusal formats must be documented in prose. The density of detail (routing mechanism, both payload shapes, cost warning) is justified, though the specific counts '5,798 across 1517 sources' add minor bulk.

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 has only one real parameter, rich annotations, and no output schema, the description fully compensates for the missing output schema by specifying both the success shape and the refusal shape with enumerated reason codes. It covers cost, usage context, sibling differentiation, and behavioral guarantees. Nothing an agent needs to decide whether to call this tool and to interpret its result is missing.

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% — the single meaningful parameter 'question' is fully documented in the schema including alias handling (query, q, prompt, text, input). The description adds contextual value by explaining that the question gets routed to auto-fill arguments across 5,798 tools, but it doesn't add parameter-level detail beyond that. Baseline 3 is appropriate since the schema carries the parametric documentation burden.

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 a specific purpose — 'Hallucination-resistant answer mode for high-stakes reads' — pairing a precise verb+resource with a clear scope. It distinguishes itself from the sibling ask_pipeworx by describing the grounded extraction mechanism and the structured success/refusal output. An agent can immediately tell what this tool does and how it differs from the plain ask_pipeworx sibling.

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?

Explicit when-to-use guidance is given: 'Use whenever an answer will be quoted, cited, or acted on, and the agent must not invent facts,' with concrete example domains (financial verdicts, legal claims, medical lookups, public statements). It also states the when-not-to-use condition directly: 'Costs one extra LLM call vs ask_pipeworx — prefer ask_pipeworx for casual lookups,' naming the alternative and the deciding factor.

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