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

Beyond annotations (destructiveHint=true, idempotentHint=false) it discloses that the render replaces the clip but the change is undoable, that it costs no Horizon Units, that it returns a job_id requiring job_status polling then preview_frame, and that unsupported account options are refused by name first. It also clarifies what is preserved ('Words, trim and crop stay'), which is exactly the non-obvious behavior an agent needs given a destructive write.

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?

It is dense but front-loaded and segmented by category, which makes 28 parameters scannable. A few fragments ('Words, trim and crop stay.', '(49 more)') and mid-token line breaks make it feel compressed rather than fully polished, but no whole sentence is wasted.

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 destructive, 28-parameter mutation tool it covers the missing pieces: what changes, what is preserved, undoability, cost, where clip_id comes from, where names come from (edit_catalog, list_clipping_presets), and the async job_id → job_status → preview_frame follow-up. An output schema exists, so return-value detail is correctly left out.

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 description coverage is 0%, so the description carries the full burden for 28 parameters — and it does, grouping them into Captions/Color/Motion/Frame and supplying numeric ranges (stroke width 0-20, filter_strength 0-100, brightness/contrast 0.5-1.5, saturation 0-2, warmth -1 to 1, effect_strength 0-1, transition_ms 60-2000, speed 0.75-1.5, fades 0-3), output_shape values, the 0 = end convention for effect_end_seconds, and the 'alone' exclusivity of template and clipping_preset.

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?

The description names a specific verb and resource ('change how a clip looks, at editor depth') and grounds it in concrete user phrasings like 'give it a cinematic color grade' and 'make it square for Instagram'. A 'Not for:' clause explicitly routes caption text/colors, music, silences, hook title, stickers, sounds and translation to named siblings (edit_clip, add_to_clip, translate_captions), so an agent can separate it from adjust_look, retime_clip and frame_clip without opening schemas.

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?

It states when to use it (styling requests at editor depth) with several representative utterances, and adds a directive constraint: 'Set only what was asked.' It also gives explicit exclusion conditions with the alternative tool for each, plus special-mode rules ('template ... alone'; 'clipping_preset ... alone, on clip_id and also_clip_ids').

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