s2_recommend_multi

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

Annotations already declare the operation read-only, idempotent, and non-destructive. The description adds useful non-obvious behavioral context: it returns a unified envelope with matching_mode similarity_ranked, and the seed IDs are echoed in query.params so the response receipt pins them. This goes beyond the annotation coverage.

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?

Two sentences, with the primary purpose front-loaded and the return-behavior detail in the second sentence. There is no filler or repetition; every sentence contributes useful 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?

Given the rich input schema, output schema, and annotations, the description covers the essential purpose and a key invocation quirk (seed IDs in query.params). It does not explain the 'unified envelope' jargon or route between siblings, but the remaining gaps are minor for an agent that has access to the full schema.

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 nested input schema has clear descriptions for positive_paper_ids, negative_paper_ids, and limit, so the schema carries the parameter documentation burden. The description adds conceptual context ('positive seed papers', 'negative ones') but no additional syntax or constraints beyond what the schema already provides.

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 clearly states the core behavior: return papers similar to positive seed papers and dissimilar to optional negative ones. It distinguishes from s2_recommend_single by emphasizing a set of positive seeds, though it does not explicitly name the sibling or say 'recommend'.

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 implies use for multi-seed similarity recommendations with optional negatives, but it gives no explicit guidance on when to choose this over s2_recommend_single or other siblings. No alternatives, exclusions, or when-not-to-use conditions are stated.

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