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

Adds rich context beyond the destructiveHint=true annotation: the run and session flip to 'cancelled', the runner interrupts the in-flight turn within seconds, the agent itself is not paused, and already-finished runs respond with their final status rather than erroring. This discloses effect, timing, and edge behavior that the annotations alone do not.

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?

Front-loaded with the verb and resource, then effects, then constraints, then usage. Every sentence carries information, though the description is on the denser side; 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 single-parameter mutation tool with no output schema, the description covers effect, timing, auth requirement, parameter provenance, and finished-run behavior, leaving no material gap for correct invocation.

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%, so baseline is 3, but the description adds provenance by telling the agent to source run_id from list_agent_runs. It adds useful meaning without detailing format (the schema already states UUID).

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+resource (cancel one live AI agent run) with clear scope, and the phrase 'AI agent run' distinguishes it from sibling cancellers like cancel_execution, cancel_data_run, and cancel_execution_step. An agent can identify the target 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?

Gives explicit when-to-use triggers ('an agent run is working on something it should not') and routes the agent to list_agent_runs to obtain run_id, plus notes admin-only restriction and the already-finished-run behavior. It stops short of naming sibling cancellation tools as explicit alternatives, so a 4 rather than 5.

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