Agency Search Tool

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

Annotations already declare readOnly, idempotent, non-destructive, closed-world, so the safety profile is covered. The description adds data-quality caveats beyond that: missing fields mean unknown, and website findings may be incomplete or outdated — real behavioral context an agent needs before trusting the result.

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?

Three tight sentences, front-loaded with the action and the ID source, then the return payload, then the caveats. No filler and every sentence carries information.

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, the description compensates by enumerating the returned fields and flagging reliability caveats. For a one-parameter read tool with full annotations, nothing material 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 coverage is 100% and documents 'Stable agency ID', so the baseline is 3. The description adds provenance ('an ID returned by search_agencies'), telling the agent where a valid value comes from rather than just its type.

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?

States a specific verb ('Retrieve') and resource ('a listed agency'), and ties it to the sibling by saying the ID comes from search_agencies. It also enumerates what is returned (services, offices, contacts, ratings, website findings), so an agent knows exactly what this call yields vs. the search tool.

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?

Clear usage context: call this with an ID obtained from search_agencies, implying the search-then-detail workflow. No explicit exclusions or when-not guidance, but with a single sibling the routing is unambiguous.

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

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

Annotations already cover the read-only, non-destructive, idempotent, closed-world profile, so the description rightly spends its words elsewhere: result ordering by editorial rating, a pagination contract, and two caveats (incomplete catalog coverage; synonyms broaden discovery but do not certify platform-specific expertise). The synonym caveat is genuinely useful semantic disclosure an agent could otherwise over-read. Return-field detail beyond total/next_offset and rate/latency behavior are absent.

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?

Four short sentences, front-loaded with the search capability and its dimensions, then ordering, then pagination mechanics, then caveats. No sentence restates structured data or pads the definition; each one carries distinct information.

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, the description compensates by naming structuredContent keys (total, next_offset) and the sort order, which is the minimum needed to page reliably. The remaining gap is the per-result shape (does a record carry the id needed to call get_agency?), which leaves the handoff to the sibling under-specified. For a five-parameter, fully annotated read tool it is otherwise complete.

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% and the property descriptions already explain q, place formats, the market caveat and limit/offset defaults in detail, so the baseline of 3 applies. The description's only added parameter meaning is restating the pagination pair and pointing at total/next_offset; it adds no new query syntax, matching behavior, or combination rules (e.g. how q and place interact).

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 (Find) plus resource (listed professional service providers) and enumerates the searchable dimensions: name, service/synonym, and office location. That is far more precise than a restatement of the name. It never names the sibling get_agency, so the search-vs-fetch distinction is left implicit rather than stated.

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?

"Use offset and limit to paginate" plus the note that structuredContent carries total and next_offset tells the agent exactly how to drive the call. The market field's schema text routes the agent to omit the filter for office-location search. There is no explicit when-not or alternative-tool steering (e.g. 'use get_agency once you have an id'), which keeps it out of 5 territory.

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