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

Annotations already cover safety (readOnly, idempotent, non-destructive), so the description is free to add genuinely new behavior: the output lands as REFERENCE.md in markdown, the 120 s window cap, both source and composition frame rates are emitted, and measured/annotated/derived/unknown values are kept separate. The provenance-separation and 'frames worth checking' notes are exactly the kind of behavioral detail annotations cannot convey.

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?

One dense sentence plus a short clause, with the core transformation front-loaded before the contents list. Every element (artifact name, window limit, frame rates, provenance rules) carries information, though the packed middle list is harder to scan than separate lines would be.

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 five optional parameters, an output schema present, and safety annotations already covering the side-effect profile, the description closes the remaining gaps an agent needs: the two input modes, the 120 s ceiling, the deliverable format, and how frames/values are reported. Nothing required for a correct call 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 60% and two parameters (scene_id, video_slug) have no schema description at all. The description compensates by making the two selection modes explicit — a single scene versus a start/end window of video up to 120 s — which the schema expresses nowhere (no anyOf or mutual exclusion), and it contextualizes target_fps as the composition frame rate distinct from the source rate.

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 transformation: a chosen scene or a <=120 s window becomes a frame-numbered reference (REFERENCE.md in markdown) for motion work. The enumerated contents (scene boundaries, text entrance/landing/hold, camera moves, audio cues) make the artifact concrete. It does not name a sibling to distinguish from get_scene or compare_to_reference, so it stops short of a 5.

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?

The two input modes ('one chosen scene, or a window of up to 120 s') imply when to use it, and the 120 s cap is a useful guardrail. However there is no explicit statement of when to prefer this over get_scene, get_frames or compare_to_reference, and no exclusions or prerequisites are given.

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