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

Annotations only declare it is a non-read-only, non-destructive, closed-world write; the description adds real behavioral context beyond that: the edit is atomic, it 'Saves a new revision', it is undoable via video_project_write action=undo, and it carries account-tier/export-limit beta constraints. It stops short of describing failure behavior (e.g. what happens when expectedRevision does not match), which for an optimistic-concurrency mutation is a notable omission.

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?

Purpose is front-loaded in the first clause and every sentence carries information (routing, revision model, beta constraints). The parenthetical-laden operation list and the trailing beta/export-limit paragraph are dense but not filler, though the beta tier details are more verbose than a caller strictly needs.

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?

No output schema exists, and the description partly fills that gap by stating the reply is a project summary with an echo=document option for the full document. Prerequisites (video_guide) and the mutation's reversibility are covered. The main residual gap is the missing explanation of revision-conflict behavior for a tool whose required inputs are a revision and an idempotency key.

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 20%, so the description has to compensate, and it partially does: it names every parameter and spells out the operation namespaces that make the otherwise untyped operations[] array usable, plus echo=document semantics. But expectedRevision (concurrency contract) and idempotencyKey (dedupe behavior) are only named, never explained, leaving the two riskiest parameters semantically opaque.

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 ('atomically apply') and resource ('typed edit batch'), and enumerates the operation namespaces it accepts (scene.*/layer.*/audio.*/project.*/brand.apply/look.*), so the agent knows exactly what this tool does. It differentiates itself from video_project_write (undo) and video_projects (read), but never contrasts itself with the closest sibling, video_edit_validate, leaving the apply-vs-validate boundary implicit.

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 concrete routing for adjacent operations ('undo with video_project_write action=undo', 'video_projects action=get reads it any time') and a prerequisite ('Call video_guide first for the catalog, the exact input schema of each action'). It does not, however, state when to apply versus when to validate first, so the most relevant alternative goes unaddressed.

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