Generate cheapest-first routing

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

Annotations already declare readOnly/idempotent/non-destructive/closed-world, and the description adds real context on top: no network calls, no files written, nothing deployed, and the user's active model is left untouched. It also discloses a genuine limitation of the generated artifact — the LiteLLM fallbacks fire on API errors only and do not enforce the quality guards or personal-data policy — which is the kind of caveat no annotation field could convey.

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?

Four sentences, each carrying distinct load: the transformation, the prerequisite and trigger, the side-effect profile, and the return shape. The primary output is front-loaded and there is no filler.

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 six-parameter tool with nested guard objects and an output schema, the description covers what is produced, what is not touched, and the key semantic gap between the routing file and the proxy config. Return values are summarized rather than duplicated from the output schema, which is the right division of labor.

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 description coverage is 100%, so every parameter (guards, pii_policy, response_format, include_unknown_cost, chain) is already documented in the schema. The description adds only conceptual framing — that ordering is cheapest-first and that guards/policy exist — without per-parameter syntax or interaction detail. Baseline 3 is appropriate when the schema does the heavy lifting.

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 states a specific transformation with named artifacts: a finished eval run (results.json) becomes a cheapest-first routing.json plus a LiteLLM proxy config. It also positions itself against the sibling modelfinder_rank_eval_results, so an agent can distinguish it without opening a 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?

It gives an explicit trigger: 'Use after modelfinder_rank_eval_results when the user wants a fallback chain.' That is clear sequencing plus a selection condition. It does not name what to do instead when the user wants something other than a fallback chain, so no explicit exclusion is present.

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