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

Beyond the readOnlyHint/openWorldHint annotations, it discloses non-obvious operational facts: a hard cap of 20 partners, CC BY 4.0 attribution, and a per-call price of $0.002 USDC settled via x402 on Base, Solana or Arc with failed calls uncharged. That payment/charging behavior is exactly the kind of detail annotations cannot express. It stops short of covering rate limits, latency, or behavior on unknown gene symbols.

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?

Purpose is front-loaded in the first clause, followed by returned fields, the cap, the scope exclusion, and cost — all useful, no filler. The final pricing/payment sentence is dense but earns its place because it affects the decision to call.

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-value explanation is optional, and annotations cover the read-only/open-world safety profile; the description still usefully names the returned fields and flags cost and the 20-partner ceiling. What's missing is guidance on invalid gene symbols and how to interpret the 0-1 confidence/evidence scores for filtering.

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 both parameters (gene, max) are already documented in the schema, including the 1-20 range and default. The description only restates the 'up to 20 partners' cap and that the gene is human, adding little beyond the structured fields. Baseline 3 applies.

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 resource (STRING protein-protein interaction partners), a specific input scope (a human gene), and the exact fields returned (partner gene name, combined confidence 0-1, evidence scores). This is clearly separable from neighboring tools like health_uniprot_protein_knowledge, health_kegg_pathway_genes and health_reactome_biological_paths, and the explicit 'No enrichment analysis' boundary reinforces that.

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?

Usage is only implied: it is the interaction-partner lookup for a human gene, and the 'No enrichment analysis' clause functions as a weak when-not. There is no explicit routing advice such as 'use uniprot for full protein annotation' or prerequisites like needing a valid HGNC symbol, so the agent must infer the alternative-tool decision itself.

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