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

Annotations already cover safety (readOnly, idempotent, closed-world, non-destructive), and the description adds context they cannot: only each prompt's latest answers, so result size scales with tracked prompts, plus the important semantic caveat that 'not exposed is never counted as none ran' for search-reporting engines. What it omits is any pagination or truncation guidance despite flagging unbounded growth, and it does not say what an empty result looks like.

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?

Front-loaded with purpose and alternatives before the field inventory, and every clause carries information. The long mid-sentence run of return fields is dense and would scan better as a list, but with no output schema that content is earning its place.

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?

No output schema exists, so the description must carry the return shape itself — and it does, naming each field with meaning (position broken into rank/of/bucket with a null case, the own vs all vs domains citation split, observed searches with counts and per-engine exposure). Combined with the sibling routing and the latest-only scoping caveat, an agent has everything needed to call it correctly and interpret the 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?

The single siteId parameter is fully documented in the schema, including the connected-app vs API-key distinction and the constraint that a supplied id must match the selected site. The description adds nothing about siteId, so the schema does all the work; with 100% coverage that is the baseline 3.

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 framing — 'the raw answers behind get_citations' — then enumerates three concrete jobs (quote what an engine said, see which pages it cited, see what it searched for). That is a clear verb+resource statement that also differentiates it from the sibling that produces verdicts rather than raw evidence.

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 names when to use it and routes to three alternatives by condition: get_citation_history for earlier checks, get_citations for the per-prompt verdict and share of voice, get_sources_report for domains ranked across all prompts. Each alternative comes with the reason that selects it, leaving nothing to inference.

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