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

Annotations cover safety (readOnly=false, idempotent=false, openWorld=true, destructive=false), and the description adds the key behavioral trait annotations cannot express: the call is asynchronous and returns before the run finishes. It stops short of stating whether duplicate concurrent runs are allowed or what quota/billing impact starting a run has.

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, action first, async caveat second, zero filler. Every clause contributes information an agent needs before invoking.

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 full schema coverage and no output schema, the definition is nearly sufficient: the agent knows it will get run info back immediately. It could be complete by pointing at how to observe the run afterward (getRun/listRuns or the webhook_url events), but that gap is minor.

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% with 5 documented parameters (including team-pinning rules, sandbox files, and webhook event types), so the schema carries the parameter burden and the description adds nothing beyond the name/schema. 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 and resource (start an agent run) plus its asynchronous nature, which cleanly separates it from read-side siblings like getRun, listRuns, and listOrgRuns and from stopRun/stopBuilderRun. It does not explicitly name a sibling, but the purpose is unmistakable.

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?

'Returns immediately - does not wait for completion' implies a fire-and-forget pattern and hints that results must be fetched separately, but no alternative tool is named (e.g. getRun/listRuns for polling) and no preconditions or when-not-to-use guidance is given. Usage is inferable, not stated.

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