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

Annotations only carry the generic readOnly/destructive/openWorld flags; the description adds the substantive behavior an agent needs: cost scaling across resolution and audio, the ~11 coins/s minimum billing for motion control, the required pairing of image_url with motion_video_url, and the async contract (returns job_id, poll check_job until status=done). That is real information beyond structured fields.

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?

Dense and front-loaded — defaults first, then the question protocol, then the special cases — with almost every sentence carrying operational payload. It runs long as a single block and partly duplicates the schema's aspect-ratio and duration enumerations, which keeps it out of the top tier.

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 9-parameter, no-output-schema, multi-model generation tool, the description covers the missing pieces: defaults, cost tradeoffs, the ask-first decision rule, model-specific parameter requirements, and the polling workflow after job_id is returned. Nothing needed to invoke it correctly is absent.

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?

With schema coverage at 56%, the description carries much of the load: it names the default model and its 480p/8s/no-audio configuration, explains what each reference parameter is for (character anchors, multi-person, style), and tells the agent to set duration to the driving clip length. A few parameters (aspect_ratio values, duration set) are still only fully specified in the schema, so it is a 4 rather than 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?

States a clear verb+resource ("Generate a video") and immediately reveals it is a multi-model dispatcher with defaults, which separates it from generate_image and check_job. The opening sentence is mostly a restatement of the tool name, so it just misses a 5, but the follow-on specifics make the purpose unambiguous.

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 prescribes agent behavior: ask the user for aspect ratio/duration/audio/resolution before calling, skip the questions when details are already given or the user asked for "the cheapest." It also routes a specific sub-case (character consistency, motion control) to the right parameters and model, leaving nothing to inference.

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