locate_evidence

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

With no annotations, the description carries the full behavioral burden and does so: no fallback to repository search, error on missing text, the silent-failure hazard ('cannot tell whether the evidence was absent or merely never searched for'), checkable spans derived from the original source, and abstention on invalid ranker output. This is unusually complete failure-mode and provenance disclosure.

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?

Purpose is front-loaded in the first sentence and every paragraph carries real content about the operational boundary. It is longer than strictly necessary and the self-referential aside ('worth stating twice') costs a sentence, but the repetition guards against a genuinely likely misreading.

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?

An output schema exists, so return-value explanation is not required (the description's mention of offsets/hashes is a bonus). However, for a 7-parameter tool with 0% schema coverage, the description is incomplete — the tuning parameters (min_score, max_matches, passage_mode, calibration_id, source_id) have no explanation anywhere, leaving an agent unable to reason about result count or filtering.

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 0% and 7 parameters exist, so the description must compensate. It explains text and query semantically (going beyond the bare schema titles), but leaves min_score, source_id, max_matches, passage_mode, and calibration_id entirely undocumented in both schema and prose — five of seven parameters are opaque.

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 opening sentence gives a specific verb ('locate') and a precise scope ('exact spans inside text without generating an answer'), and the body explicitly distinguishes this from repository/workspace retrieval ('does not read the workspace and has no index behind it'). An agent can tell it apart from entroly_retrieve and smart_read without opening any schema.

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 states when to use it (pass content you already hold, typically from read_source_file), what the query does and does not do (ranks spans within the text; does not search for the text), and an explicit when-not/error case (calling it without text is an error, not a whole-repository query). The boundary conditions are called out rather than left to inference.

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