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

Annotations only declare readOnly/openWorld/idempotent hints, so the description carries the burden — and delivers. It discloses the refusal mechanism with an enumerated set of refusal reasons (not_in_source, no_tool_match, tool_error, data_truncated, llm_error), the exact success and failure return shapes, and the extra LLM call cost. No contradiction with annotations; readOnlyHint=true is consistent with a read-only grounded-answer tool.

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 every sentence earns its place: purpose, mechanism, success/failure formats, use-case triggers, and a cost comparison against the sibling. The most important differentiator (grounded/hallucination-resistant) is front-loaded. Minor redundancy: the refusal behavior is described twice, once as return format detail and once in the refusal_reason list.

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 routing tool with no output schema, the description is remarkably complete: it covers the success return shape, all five refusal reasons, the extraction mechanism, and the cost trade-off. The only gaps are minor — the semantics of the 'confidence' field and the exact shape of 'evidence' are not defined, and there is no guidance on how to relay a refusal to the user.

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%, so the schema already documents the single 'question' parameter with aliases (query, q, prompt, text, input). The description adds no parameter-specific semantics beyond the phrase 'fills arguments,' which is behavioral context about internal routing rather than parameter guidance. Baseline 3 applies.

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.' It names the sibling it parallels ('Same routing as ask_pipeworx') and states its distinguishing behavior — extracting the answer using ONLY the tool result — so an agent can tell it apart from ask_pipeworx and ask_pipeworx_beta 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?

Explicitly states when to use ('whenever an answer will be quoted, cited, or acted on... financial verdicts, legal claims, medical lookups, public statements') and when not to ('prefer ask_pipeworx for casual lookups'), including the cost trade-off ('costs one extra LLM call'). This leaves no ambiguity about tool selection.

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