run_eval

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

No annotations exist, so description carries full burden. Discloses mutating nature (POST), cost (~$0.01/run), scoring algorithm, and return behavior (summary with scoredCount and reason). Lacks mention of idempotency handling or full response structure, but adequate.

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?

Single paragraph, dense but efficient. Front-loaded with the main action and endpoint. Could benefit from bullet points for preconditions, but still clear and concise.

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 5 parameters, no output schema, and important preconditions, the description covers almost everything: failure modes, costing, preconditions, and partial return structure. Missing full response shape but sufficient for agent to invoke correctly.

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% with good descriptions. The tool description adds context linking parameters to behavior (e.g., recentCount defines how many calls, label as substring filter, idempotencyKey for dedup). Adds value beyond 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?

Clearly states the verb (Start a new eval run immediately), resource (eval-runs via POST), and what it does (scores recent calls against default and custom criteria). Distinct from siblings like compare_eval_runs or run_eval_dataset.

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?

Provides explicit conditions: Pro+ only, needs OPENAI_API_KEY, requires plaintext storage opt-in. Gives failure modes (403, 500, zero scored). Does not explicitly contrast with run_eval_dataset, but context is strong.

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