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

The description adds genuine context beyond annotations: the project is not mutated, the result is images only, and 'sampled stills do not prove smooth playback or audio' is a valuable expectation-setter, plus a per-turn cap that explains why readOnlyHint is false. However it never reconciles with destructiveHint=true and openWorldHint=true — an agent reading both signals cannot tell what is actually destroyed or what external state is touched, which is the one thing that most needs disclosure here.

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?

Front-loaded with purpose and non-mutation, then mode rules, then the output caveat and limit — a sensible order with little filler. The prose is dense with domain jargon ('candidate finish', 'programs stack', 'source composition clock') and the instance-defaults clause is a run-on, which costs readability but not correctness.

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?

An output schema exists, so the description need not enumerate return values, yet it still flags the return shape (images) and its limits. Combined with the mode rules, the instance defaults and the per-turn cap, an agent has enough to call it correctly; what is missing is failure behavior, notably what a stale expected_spec_hash produces.

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?

At 50% schema coverage with an untyped 'instance' object (additionalProperties: true), the description carries real weight: expected_spec_hash must be current, programs must be the complete candidate stack, and instance needs only id, role and its pinned revision reference with version, enabled, seed, clock, assets and params defaulting. That is meaning the schema cannot supply. It leaves read_skills, request_key and publication_proposal_id untouched, so not a 5.

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 and resource — render one to three actual output frames of a candidate finish on this exact Film picture — and the key qualifier 'without applying it' separates it from apply_visual_program and from build/read_visual_program_* siblings. An agent can tell what this does and what it does not do without opening the schema.

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 a real usage rule for the instance path ('for a generated clip/overlay use instance and an empty programs array; frames then use the source composition clock') and a hard budget ('three inspections per turn'). It does not explicitly route the agent between inspecting versus applying, or say when inspection is unnecessary, so it stops short of full alternative guidance.

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