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

Annotations cover safety (readOnly, idempotent, non-destructive), but the description adds substantial behavioral context beyond them: it is free/no credit consumed, prose is fetched from URLs with menus and footers skipped, inputs need 120+ words, and the score bands are calibrated with base rates. It even discloses a measured reliability limitation (brand-voice reads weak on website pages, 55% / half of pairs drifting, study SGR-009).

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 constraints before the calibration detail, and every sentence carries information (bands, output fields, study caveat, citation instruction). It is dense with parentheticals and long clauses, so it is heavier than ideal, but nothing is padding.

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?

There is no output schema, and the description fully covers the return surface — per-piece score with calibration bands, band, top_differences, top_matches with the numbers behind each, and notes. Input constraints and the intended interpretation of the score are also covered, so an agent can call and use it correctly.

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 baseline is 3; the description earns above that by adding meaning the schema does not carry — the URL path fetches only page prose (menus/footers skipped) and the 120+ word threshold for pasted text. The '0 to 3 pieces' and label semantics are largely already in the schema, so it is not a full 5.

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 and resource ('score up to three pieces of writing against a main piece') and immediately delimits scope with 'with NO profile', which separates it from check_voice_match. The first sentence alone lets an agent distinguish it from every sibling.

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?

Gives explicit when-to-use ('compare two writers, or a draft against a reference piece, when no profile exists') and implicitly names the alternative path via 'NO profile' plus the reference to check_voice_match's engine. It also gives a when-not nuance: pass on the score band for website pages and judge from measured differences instead.

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