Start Control Build

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 does so: credits are consumed only when confirm_estimate=True, the model is locked until publish/fail/discard, other writers are refused meanwhile, and reads show the last published controls. It also enumerates refusal codes (no_proposal, review_stale, generation_active, 402 balance) as structured returns, which an agent could not infer from the schema.

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?

The critical decision (confirm_estimate gating cost and mutation) is front-loaded in the first sentence, and the numbered two-step flow is easy to scan. The refusal-code list is long but each entry adds actionable recovery information, so little is wasted, though the description is heavier than strictly necessary.

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 two-phase, credit-consuming mutation that locks a model, the description supplies the estimate/confirm loop, the locking semantics, alternate proposers, and the polling path to terminal state. With an output schema present it need not restate return shapes, yet it still documents them concretely, leaving no gap an agent must guess at.

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 80% and the schema already documents each parameter, so the baseline is 3; the description goes further by tying model_version and set_revision to the values returned in the proposal and explaining that confirm_estimate=False starts nothing and charges nothing. It does not explain server_version or model_id beyond the schema, keeping this short of a 5.

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?

Opens with a precise verb+resource ('Start the model's proposed control build') and immediately scopes it against siblings by explaining that generate/refine/edit/regenerate_controls only PROPOSE and that this tool is what actually starts a build. An agent can distinguish it from regenerate_controls or resume_control_generation without opening any 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 an explicit ordered workflow: first call with confirm_estimate=False to get the estimate, then call with confirm_estimate=True plus model_version and set_revision after user review. It also names the alternatives that merely propose and the poll target (get_control_generation_status), so when-to-use and when-not-to-use are both covered.

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