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

Annotations already declare destructiveHint=true, but the description adds real value beyond that: it enumerates what gets destroyed (workflows, jobs, attempts) and confirms the action is recoverable via retry. It does not say whether cancelling a finished run errors or is a no-op, which is the remaining gap.

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 compact sentences with the mutation warning front-loaded, then the scope, then the recovery pointer. No filler and nothing buried.

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 one-parameter mutation with no output schema, the description covers safety intent, destruction scope, and the follow-up recovery tool. Nothing an agent needs to call this correctly is missing.

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% and the single run_id parameter is already documented in the schema, including its provenance ('from depot_ci_list_runs'). The description adds no syntax or format detail beyond that, so the 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 verb (cancel) and resource (Depot CI run) plus the exact blast radius: 'all of its unfinished workflows, jobs and attempts.' This lets an agent distinguish it from read-oriented siblings like depot_ci_get_run_status without opening the 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?

The 'WRITE — stops work in progress' lead-in sets the context for when this tool applies, and it routes the recovery path to a named sibling (depot_ci_retry_failed_jobs). It stops short of explicit when-not guidance, such as behavior on an already-completed run.

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