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

Exceptional disclosure beyond annotations: runtime expectations ('up to a few minutes' locally, 3-4 minutes online), the asynchronous start-then-poll protocol, quota consumption per day, that a cached piece re-answers without spending a check, that a modified piece is reported as such, and that a failed check refunds the day's check. This fully explains why the tool is not read-only and not idempotent, matching the annotations.

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?

Information density is high and little is filler, but the text is a wall of run-on sentences with ornamental phrasing ('drawn by the real engine', 'words that may claim something', 'a bare ground or a jump'). The single most important operational fact — that online calls are asynchronous and must be re-polled — is buried mid-paragraph rather than front-loaded.

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 no output schema, the description carries the burden of describing returns and does so at a high level ('answers what is wrong in sentences, with a picture of it at work'), plus the status strings an agent will see while a check runs and after a failure. A more explicit statement of the verdict payload shape would close the remaining gap, but an agent can call this correctly from what is given.

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, and the description earns above that by unpacking 'again: true' with real semantics the schema only hints at: it starts a new check even when cached, stale, or failed, it costs one of the day's checks, and it must not be used while a run for that version is in flight. The 'id' parameter is left to the schema, which already points to search_library { mine: true }.

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?

The description states a specific verb and resource ('Pass one of the user's pieces through every gate the pieces of the library pass') and enumerates the actual checks: shader compiles, weight, scene robustness, no dark/light flashing, brand colours, four content formats. It is clear what the tool does despite the metaphorical framing, though it never names check_film or explains the piece-vs-film boundary for an agent choosing between them.

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 concrete operational guidance for the repeat-call workflow: online the first call returns immediately, call again a few minutes later for the verdict, and a cached verdict reappears for an unchanged piece. It also states the one exclusion for 'again' ('never while one of this version runs'). It stops short of saying when to run a check at all (e.g. after create_piece/edit_film) or why to prefer this over preview_film.

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