check_recall

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

With no annotations, the description carries real weight and delivers: it discloses batched esearch intersection logic, a 50-per-batch cap to avoid NCBI 414 URL errors, deduplication, a 200-PMID ceiling, and that missing items come back with title and existence status. It also states that query warnings are passed through verbatim with a prominent alert when the term is ignored or only partially applied. It does not discuss the cost or rate-limit implications of issuing several batched requests, which is the main remaining behavioral gap.

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 text is dense but front-loaded: the core purpose comes first, then constraints, then interpretation caveats, then warning behavior. Every sentence adds information, though the phrasing is terse and jargon-heavy enough that an unfamiliar agent may need to parse it carefully.

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 the description carries no burden to explain return values, and it still usefully notes that missing entries include titles. Given the three-parameter surface, the remaining omission is the meaning of as_json and the accepted syntax for term.

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%, so the description must compensate. It documents pmids well (comma/space separated, deduplicated, max 200) and implies that term is a query with concept blocks, but it never states what query syntax term accepts. The as_json parameter gets no mention at all, leaving a third of the parameters unexplained.

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 precise statement of what the tool does: test which of a batch of known-relevant PMIDs are retrieved by a given term. That is a specific verb-and-resource pair, clearly distinct from a generic search or count tool, so an agent can identify it without opening the 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?

The description gives clear usage context and an explicit scope caveat: it only proves recall against this known set, not overall recall, and warns that the gold-standard set itself may be biased. It also points to count("concept AND pmid[uid]") for block-level diagnosis. It stops short of naming sibling tools like search or count as direct alternatives, so it is strong but not a full when-to-use/when-not-to-use mapping.

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