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

No annotations are provided, so the description carries the full burden, and it does disclose the key behaviors: the verdict can be 'unknown' when protocol or peer evidence is missing, and the comparison is scoped to identical experimental conditions. It also gives a cost signal (8 credits) and presents the tool as a query operation. It doesn't explicitly state that the operation is read-only, or explain latency or failure modes beyond the unknown verdict.

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?

Three short sentences plus a pricing tag, with the core question front-loaded so the agent understands the tool's purpose immediately. Every sentence adds information: the decision, the returned evidence, the failure mode, and the cost. Nothing is redundant.

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?

There is no output schema, so the description must explain the return values, and it does: the current leader, who has beaten it since, and an unknown verdict. Cost transparency is a bonus. Minor gaps remain around the exact shape of the returned leader/beaten-by data and what happens with malformed inputs, but overall it is sufficiently complete for a 5-parameter lookup tool.

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 only 20% (only 'condition' is documented in the schema), so the description needs to compensate. It does go beyond the schema by grouping the inputs into two invocation modes and by reinforcing that 'condition' must match an exact dataset_leaderboard protocol key. However, it doesn't explain what dataset_key, metric, or value mean or how they should be formatted, so half the parameters remain somewhat opaque.

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 pointed question that states exactly what the tool decides: whether a paper, or a dataset+metric+value, is state of the art. It also commits to a specific return shape (current leader plus who has beaten it under the same conditions). It doesn't name sibling tools (new_sota, sota_alarm, sota_changelog) to explicitly distinguish itself from them, so it falls short of a 5.

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 clearly signals the two calling contexts that correspond to the schema's anyOf: pass a paper_id, or pass a dataset_key+metric+value (plus condition). The fallback rule, that missing protocol evidence or peers yields an unknown verdict, tells the agent what to expect when those inputs are incomplete. There is no explicit when-not guidance or a direct comparison to alternatives such as new_sota or sota_alarm.

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