manage_eval

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

Annotations already declare destructiveHint=true, idempotentHint=true and openWorldHint=true, so the safety profile is structured. The description does add value by warning about confirmation and by scoping cancel to queued builds (not running ones), but it says nothing about what happens to running builds during restart, whether restart is bounded, or the effect on the queue ordering.

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, front-loaded with the action set, followed by the one safety-critical caveat. Nothing redundant; every clause carries 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?

With annotations covering safety and an output schema present, the description need not explain returns. It covers actions and the destructive-cancel warning, leaving only sibling differentiation and the running-vs-queued build nuance under-specified.

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 only 33% (only 'instance' is described), so the description must carry the action semantics. It does partially: the four enum values map to restart-failed, restart-aborted, cancel queued builds, and bump to front. But it doesn't explain how 'instance' interacts with the default, nor whether restart applies to the whole eval or specific builds.

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 specific verbs and the resource scope: restart an evaluation's failed/aborted builds, cancel queued builds, or bump them. It is clear what the tool does, but it never distinguishes itself from the per-build siblings restart_build, cancel_build and bump_build, so an agent must infer that this is the evaluation-level bulk variant.

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 'confirm with the user first' instruction gives real guidance for the cancel action, and the action list implies the four usage modes. But there is no statement of when to pick manage_eval over restart_build/cancel_build/bump_build, which is the main selection decision an agent faces here.

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