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

Annotations only say it is destructive, non-idempotent, and open-world; the description adds the materially useful facts an agent cannot get from those flags — it spends team credits, the price is the one after_steps lists (voiced price for a voiced language), a mismatched figure is refused, and credits are refunded if it fails. It also discloses the asynchronous job pattern and the polling interval, which is exactly the behavioral context the hints do not carry.

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?

Four sentences, front-loaded with the action and the two variant behaviors before cost and follow-up. Some phrasing is compressed to the point of awkwardness ('Answers an id at once'), but no sentence is filler — each introduces pricing, eligibility, or the polling loop.

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 correctly takes on the return contract (an id plus how to poll it) and covers cost, refund, and language eligibility for a credit-spending mutation. The only soft spot is that 'part' — which sub-video is transformed — is never touched on, leaving the agent to rely on the schema alone.

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 the baseline is 3, but the description adds real meaning: credits must equal the after_steps figure (a different number is refused) and language must be one after_steps offers, both cross-tool constraints the enum and integer schema cannot express. The polling sentence also tells the agent what the returned id is for, though 'part' is left entirely to the schema.

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 ('make one finished video again in another language, into a new video beside it') and immediately distinguishes the two output modes: a UGC video gets its own script and presenter, a product video keeps its picture and re-records the voiceover. That is enough for an agent to separate it from make_video, change_video, make_longer, or other_sizes 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 a hard precondition ('Only in a language that after_steps lists for it') and clear next-step guidance: the call answers an id at once, then poll get_video (or get_product_video) about once a minute until status is ready or failed. It does not explicitly compare itself to sibling tools that also transform an existing video, so the when-not side is inferred rather than stated.

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