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

Annotations already cover safety (readOnlyHint, openWorldHint, idempotentHint, non-destructive), and the description adds rich behavior beyond them: the full refusal contract with five enumerated refusal_reason values, the verbatim-evidence return field, and the extra-LLM-call cost. The description aligns perfectly with the annotations — it describes a read/analyze operation with no destructive or mutating behavior.

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?

Five sentences, each with a distinct job: purpose, routing mechanism, success/refusal return contract, when-to-use guidance, and cost/alternative tradeoff. Since there is no output schema, the detailed payload documentation earns its place rather than being fluff. It is front-loaded with the key differentiator ('hallucination-resistant') before any mechanics.

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?

With no output schema present, the description correctly absorbs the return contract, enumerating the success shape {answer, evidence, confidence, source, fetched_at, refusal_reason:null} and all five refusal_reason values. Combined with annotations covering the safety profile and explicit differentiation from ask_pipeworx, an agent has everything needed to invoke correctly and interpret every possible 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 coverage is 100% and all five aliases reduce to the 'question' property, so the schema carries the parameter documentation. The description adds processing semantics beyond that: the question is routed through tool selection, argument filling, and data fetching, and answers are extracted only from the tool result — implying agents should phrase questions answerable from source data. Marginal but real value over the schema alone.

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?

Opens with a specific, differentiated purpose — 'Hallucination-resistant answer mode for high-stakes reads' — and states the mechanism: routes across 5,798 tools / 1,517 sources, then EXTRACTS the answer using ONLY the tool result content. It explicitly contrasts with the sibling ask_pipeworx ('Same routing as ask_pipeworx... prefer ask_pipeworx for casual lookups'), so an agent can distinguish it from near-names without inspecting schemas.

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?

Gives an explicit positive condition — 'Use whenever an answer will be quoted, cited, or acted on' — with concrete high-stakes domains (financial verdicts, legal claims, medical lookups, public statements). The negative condition is also explicit: 'prefer ask_pipeworx for casual lookups,' naming the exact alternative and the cost reason. Nothing is left to inference.

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