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

Well beyond the annotations: it discloses that processing runs on a hosted GPU, consumes credits, is cached per media/stretch/mode, and returns a specific artifact (grayscale MP4 or PNG with white = subject). Annotations only say readOnly=false, destructive=false, so this context is genuinely additive. The notable gap is execution model — nothing says whether the call blocks or enqueues a job (relevant given cancel_job/job_status siblings), and the 5-minute cap lives only in the schema.

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?

Three tight clauses with zero filler: purpose, cost/model, output format and color semantics are all front-loaded. Every sentence fragment earns its place and the output contract lands before any optional detail.

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 no output schema, the description does the needed work on return values (file type per input, grayscale encoding, white = subject). What's missing for a 6-parameter, credit-consuming, non-read-only tool is the async/blocking behavior and how the resulting matte is retrieved or reused.

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 all six parameters are documented, and the description adds only oblique meaning via 'cached per media, stretch and mode' — hinting that from/to and fast/occlusion affect caching without naming them. That is marginally useful but doesn't explain the soft-vs-hard-matte tradeoff beyond what the schema's own param descriptions say.

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?

Precise verb+resource: separates subject from background, with the meaningful split between video mattes (people) and image mattes (main subject). It does not name or compare against any sibling, but no sibling in the list overlaps with matting, so an agent can still place it unambiguously.

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?

Usage is implied rather than stated: 'video mattes find people, image mattes the main subject' tells the agent what each input type yields, and the credit-cost note hints at a tradeoff. There is no explicit when-to-use/when-not, and no pointer to alternatives such as grab_frame or filmstrip for cheaper previews.

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