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

Annotations already declare readOnlyHint=false, destructiveHint=false, idempotentHint=true and openWorldHint=true, so the core safety profile is pre-declared. The description then adds genuinely non-obvious behavior: credits are held and refunded on failure, a fixed server suffix is appended to the prompt, the prompt is passed verbatim, trial/concurrency/daily-cap limits, and async polling every 20 s via autoshort_get_video. That is well beyond what the annotations cover, so a 3 understates the description's contribution — it would score higher if the omission bar did not already discount annotation-covered facts.

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 purpose and the use/don't-use routing, then moves through pricing, policy, optional params, limits and polling. It is dense and largely earned, though it reads as a single long block and the pricing/trial-limit details are repeated in places, which costs it one point.

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 return-value burden and does: it echoes promptSent and the seed, and explains that the asynchronous result comes from polling autoshort_get_video until COMPLETED (videoUrl MP4) or FAILED (errorCode, creditsRefunded). Combined with cost, policy, and limit disclosures, an agent has everything needed to invoke and follow up correctly.

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 semantics the schema does not: prompt-wins-over-templateId precedence, templateId pulls a ready prompt/seconds/aspect ratio and an unknown id is INVALID_INPUT with no charge, character_id starts the clip from the sheet's chosen image but does not guarantee an identical face and yields CHARACTER_NOT_FOUND when not owned/ready, plus the seed echo for repeatable takes. These are decision-relevant rules an agent could not infer from the schema alone.

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 ('Turn a text-to-video prompt into ONE AI motion clip'), names the model (Seedance 1.5 Pro), and explicitly distinguishes itself from autoshort_generate_chat_video. An agent can route between the two video-generation siblings without opening either 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 an explicit positive trigger (user pastes a cinematic/video-model prompt with camera, lens, lighting, aspect-ratio syntax) and an explicit negative route ('For a narrated story or facts video use autoshort_generate_chat_video'). It also states a mandatory prerequisite workflow: call autoshort_quote_video first and quote the price before creating.

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