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

Goes well beyond the annotations (readOnly=false, idempotent=true, destructive=false) by explaining update-in-place semantics ("One rating per cook per recipe: calling again updates it"), preservation rules ("a comment left out is kept"), the refusal case, the 2,000-character combined limit, and that made_it completes a cook and may trigger a pantry deduction. The null-vs-omitted distinction for optional answers is exactly the kind of contract the schema can't express.

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, then triggers, then invariants, then parameter contracts — a sensible order with no filler sentences. It is dense and rule-heavy; a few of the trailing constraints could be grouped more crisply, but every sentence carries operative 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?

For a 10-parameter mutating tool it covers the full contract an agent needs: required inputs, authorization refusal, idempotency, omission vs clearing, size limits, and the cook-completion side effect. Return values are out of scope because an output schema exists.

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?

With only 30% schema description coverage across 10 parameters, the description compensates well: it clarifies stars must be explicit, comment preservation on omission, made_it semantics, content_token's source and prerequisite step context, and clearing behavior via null/empty arrays. It leaves `name` and `would_make_again`/`substitutions_made` enum values to the schema, which is reasonable since the schema is explicit there.

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 verb+resource pair ("Post the cook's star rating (1 to 5) and comment") plus the side effect of marking a cook as made. It also names a sibling (get_my_recipe_feedback) as the source of content_token, so the agent can distinguish this write tool from the read-side feedback 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?

Gives concrete trigger utterances ("that was great, five stars", "I made the chili, a bit too salty"), an explicit guard against inferring data ("Keep stars explicit; never infer completion or preferences from notes"), a fallback instruction ("ask for the stars if they didn't give any"), and an exclusion ("Refuses the cook's own recipe").

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