review_cuts

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

Annotations only declare readOnlyHint=false, destructiveHint=false and openWorldHint=false; the description adds substantial context on top: it writes short preview clips and a local report (the reason it is not read-only), returns a durable job ID rather than results, and otherwise leaves the edit untouched. It also qualifies the output ('review cues, not a verdict on naturalness') and notes that renderPreviews=false skips audio. It does not state cost, runtime, or required permissions, so a 4 rather than a 5.

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?

The purpose and the sibling hand-off (job_get) are front-loaded, which is good, but the body is eight dense sentences mixing detection logic, output artifacts, caveats, pagination and mode constraints in no obvious priority order. The mode sentence ('the engine owns timing and main clips') is jargon-heavy and does little work for a first-time caller.

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 six parameters, no output schema and a job-based async return, the description does the necessary work: it names the return value, points at job_get for retrieval, and discloses the written artifacts. The remaining gap is the unexplained contextSeconds parameter and an unclear 'layer mode' prerequisite, but nothing an agent needs in order to invoke the tool is missing.

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 50%: renderPreviews, projectDir and expectedRevision are documented in the schema itself. The description adds meaning only for offset/limit (batching long edits); contextSeconds, a non-obvious windowing parameter, is never explained despite governing what audio/transcript context is examined near a join. That is roughly the baseline for a mid-coverage schema, not compensation.

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 first sentence gives a concrete verb and resource ('Review source-linked timeline joins') and immediately scopes it with 'without changing the edit', which separates it from edit_apply and preview_render. It does not name a sibling by name (aside from job_get as a follow-up), and 'joins' combined with the cryptic mode sentence leaves some room for interpretation, so it falls short of a 5.

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?

The description supplies real usage context: it routes the agent to job_get for results, tells it to use offset/limit to batch long edits, and states a precondition ('Independent standalone compositions require adaptation to layer mode'). What it lacks is explicit when-NOT-to-use guidance, e.g. how it differs from project_validate or preview_render for a reviewer deciding between them.

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