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

With no annotations supplied, the description carries the full burden and does so: it discloses the cost model (free mistake checks capped at 2 per shot, 1 free change per minute, then 3D pack minutes at 5 changes per minute), the asynchronous latency ('takes a couple of minutes'), and the blocking answer state with its two resolutions. These are exactly the traits an agent needs before mutating a film.

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 practically every clause carries non-redundant information about modes, quotas, or state transitions. It is dense and parenthetical-heavy, which makes it slower to parse than ideal, but nothing is filler.

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 6-parameter, no-annotation, no-output-schema mutation tool, the description covers prerequisites, per-mode behavior, quota/pricing, latency, blocking states, and the downstream calls that finalize the result. An agent could invoke and recover from this tool without additional context.

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 beyond the schema by explaining the consequence of each mode enum value ('keep the picture, animate again' vs 'a new picture of the same characters and place') and the billing difference between reason=mistake and reason=change. It says less about shot, change, and video_id, but those are adequately documented in 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?

Opens with a specific verb and resource: 'Remake one shot of a finished 3D film', then enumerates the three modes with what each preserves or regenerates. This distinguishes it clearly from siblings like regenerate_scene_image or animate_scenes, which operate on images or scene drafts rather than a finished 3D film shot.

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 gates usage ('Only on the user's own request'), explains the mistake-vs-change decision tree including the free-quota ordering, and describes the follow-up path via list_3d_shots, keep_3d_shot and update_3d_film. Even the edge case (a failed mistake check leaving the shot awaiting an answer) routes the agent to a concrete next call.

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