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

Annotations already declare readOnlyHint, idempotentHint, non-destructive and closed-world, so the safety profile is covered. The description adds genuinely new behavior: it costs no QuestionPunk AI credits, and it returns a prompt meant to be executed by the caller rather than executed server-side — the key mental model an agent needs. It stops short of noting failure modes (e.g. a skill id that doesn't exist for that survey), which keeps it below 5.

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 the action, then cost, then the sibling comparison, then the edge case — a sensible priority order. Every sentence carries distinct information; nothing is restated from the annotations or schema.

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 an output schema present, return format needs no explanation, and full parameter descriptions plus rich annotations carry the structural load. The description covers the remaining gaps — cost model, execution model, sibling choice, and the one conditional parameter case — so an agent can invoke it correctly on the first try.

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 still adds value by flagging that segment-comparison specifically requires segmentQuestionId and that it comes from skill_list's segmentQuestions, which is a cross-tool dependency the schema field does not spell out as clearly. The skillId and surveyId semantics are left entirely to the schema.

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?

Specific verb and resource: "Get an analysis skill's prompt, ready for one survey" — it states what is retrieved and the scope (single survey). It then distinguishes itself from the closest sibling by declaring itself "the default way to run a skill" versus skill_run, so an agent can choose without opening either 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?

Explicit routing rule: use this by default, use skill_run only when the user asks for QuestionPunk's AI to run it. It also prescribes the follow-on workflow (read answers with responses_read, then answer the prompt) and names a conditional prerequisite for segment-comparison.

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