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

Annotations only cover the safety profile (readOnly=false, destructive=false, openWorld=true, idempotent=true); the description adds the materially important context they omit: credit consumption (250 per scene), a multi-minute runtime, and the auth/sign-in requirement. No contradiction with the annotations — the explicit user-consent gate is consistent with a non-idempotent-perception but idempotent-flagged start operation.

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 core action and the cost/consent gate in the second clause, then exclusions last — a sensible ordering. It is a dense multi-sentence block, but each sentence carries distinct information (action, gating, cost, polling, auth, scope, exclusions); only the trailing exclusion list runs slightly long.

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?

An output schema exists so return values need no explanation, and the description still supplies the operationally critical facts an agent needs: cost, runtime, how to poll, auth prerequisite, and the exact class of ads it applies to. Nothing needed to invoke it correctly is missing.

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 90% and six parameters carry enums with their own descriptions, so the schema does the heavy lifting and the baseline is 3. The description's only param-adjacent value is tying cost to scenes ('250 per scene'), which loosely connects to the duration→scene-count mapping already documented in the schema, so it does not meaningfully exceed the baseline.

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 concrete verb+resource — 'Start the AI production of a prepared UGC video ad' — and immediately scopes it to a new short AI video ad in the signed-in account. The pipeline description (strategy, script, keyframes, scene videos, final cut) makes the output unambiguous and clearly distinct from prepare_ugc_video_ad and get_ugc_video_ad.

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?

Explicitly sets the gating condition ('only after the user explicitly asked to generate'), names the alternative for progress tracking ('poll with get_ugc_video_ad'), and lists concrete exclusions (editing uploaded video, captions, real human creators, stills, logos, non-ad videos). This is as close to a complete when/when-not/alternatives map as a description gets.

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