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

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint, covering the safety profile. The description adds valuable behavioral nuance: the interests are inferred (not typed), the coverage field indicates reliability ('lite' means a new account), and it clarifies that evidence itself is not returned. This enriches the agent's understanding beyond the annotations without contradicting them.

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?

The description is about 100 words, front-loaded with the core purpose, then logically flows through output details, usage guidance, and a caution. Every sentence adds value—no filler or repetition. The structure is easy to parse, with clear stopping points for an agent.

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 zero-parameter, read-only inference tool, the description covers the essential aspects: what it returns, how to interpret reliability, when to use it, and a usability caveat. An agent can confidently decide when to call it and how to act on the result. The lack of an output schema is mitigated by the detailed field descriptions.

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 zero parameters, the schema provides no information, so the 0-param baseline is 4. The description goes further by explaining the output structure (strength 0-1, active flag, coverage field) and the meaning of 'lite'. Since there is no output schema, this textual explanation is essential for interpreting the tool's results, fully compensating for the missing 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?

The description opens with a specific verb and resource ('Read what this user is actually interested in') and immediately distinguishes itself from user-typed input by listing data sources (bookmarks, labels, questions, clicks). It clearly states what is returned (topics, strength, activity, coverage), making it distinct from sibling get_* tools like get_bookmark or get_library_overview.

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?

The description explicitly provides use cases ('prioritise, filter or explain') and instructs to confirm with the user rather than asserting the result as fact. It also warns about the 'lite' coverage meaning low evidence. However, it does not explicitly name alternative tools or state when not to use it, though the 'not from anything they typed' implies it isn't for direct user queries.

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