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

Annotations only say destructiveHint=true; the description goes much further, disclosing that real credits are spent, that the server enforces the cap before every render, that failures refund themselves, and that stopping preserves completed work. These are the exact risk facts an agent needs before invoking an irreversible spend.

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 credit-spend warning is front-loaded in caps immediately after the opening clause, so the highest-risk fact is seen first. The remaining clauses are dense but each carries a distinct behavioral fact; the run-on punctuation in the rung clause is the only rough edge.

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 destructive, 5-parameter, no-output-schema tool, the description covers preconditions, cap enforcement, failure refunds, and interruption semantics. What it omits is what a successful call returns (a run identifier or status) and how to observe progress afterward, which run_control/get_status would handle.

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 60%, and the description fills the main gap by decoding the rung enum (1 every step · 2 big decisions · 3 autopilot), which the schema provides no descriptions for. It also reinforces cap_credits as human-approved and never invented, and links quote_id to the quote step. project_id remains unexplained in both places.

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?

'Start the produced run' gives a specific verb and resource, and the title plus the credits sentence make clear this is the spend-triggering execution step. It is distinguishable from quote and apply_plan, though the word 'produced' is slightly opaque without familiarity with the pipeline.

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?

It gives a concrete precondition (cap_credits must be a number the human explicitly approved in chat after seeing the quote) and spells out rung's three modes. It also explains exactly when confirm_length applies (only after a refusal citing that question), which is strong conditional guidance. It never names a sibling tool as the alternative, so it stops short of a 5.

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