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

Annotations are sparse (idempotentHint=true, destructiveHint=false), so the description carries the burden and delivers richly: PROJECT-SCOPED execution, SOURCE-time seconds that persist through later cuts, automatic re-anchoring of music/overlays/zooms/sfx, audio pitch preservation, frame-duplication stepping below 0.6x, and disclosed REPLACE-onspan-overlap semantics. Nothing contradicts the annotations; the idempotentHint is consistent with a replace-on-overlap speed setting, and the replacement is disclosed rather than hidden.

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 but lean: scope is front-loaded in caps, and every sentence delivers a distinct fact (source-time semantics, factor semantics, audio behavior, quality warning, replacement, use case). There is minor redundancy with the schema's project_id description, but no filler. The length is justified by the number of counterintuitive behaviors an agent must know before calling.

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 5-parameter mutation tool with no output schema, the description covers the essentials: execution scope, input semantics, factor constraints, side effects (re-anchoring, replacement), and even the return shape ('returns the project identity with its result'). Minor gaps remain — behavior on invalid ranges (start >= end) and how motion_motif interacts with the speed event — but motion_motif is documented in the schema, so the total package is nearly complete.

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 only 40%, and the description compensates for the undocumented start/end/factor: it defines start/end as SOURCE seconds and gives factor a full semantic spec (0.25-4.0 range, 2.0=double, 0.5=half, pitch preservation, preferred band). This adds meaning the bare 'number' types in the schema omit. project_id and motion_motif are left to the schema, which already documents them adequately.

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 states a specific verb+resource+scope: 'Speed up or slow down a SOURCE-time range of the main video', which distinguishes it from siblings like set_volume (audio) and remove_speed (removal). It closes with 'THE tool for "speed up the boring part" / "slow-mo that moment"', giving an agent a crisp retrieval cue. The analogy 'like set_volume' reinforces what kind of operation it is without conflating the two.

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?

Provides clear usage context: natural-language intents ('speed up the boring part', 'slow-mo that moment') and concrete quality guidance ('prefer 0.6-0.8x' slow motion, warned visual stepping below 0.6x). It names set_volume as a behavioral analog, which helps calibration. However, it does not explicitly state when-not-to-use or name mutually exclusive alternatives (e.g., remove_speed for undo), so it stops short of a full exclusion contract.

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