select_best_takes

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

With no annotations at all, the description carries the full burden and does well: it discloses the scoring signals (cut calls, cue confidence, retake phrases, short takes, later-take bias), the fact that it can write a selects sequence with markers, and that it shares the rough_cut_by_cues cache. It stops short of saying what happens to the project when sequenceName is omitted versus supplied beyond that hint, and gives no sense of runtime cost or failure modes beyond the schema's allowOriginal note.

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-loads the verb and resource, then enumerates scoring factors compactly in a single dense clause list, followed by return shape and the optional sequence side effect. It is information-dense rather than padded, though the scoring-factor list is long enough that a reader must parse carefully.

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 compensates by describing the return form (ranking with per-file reasons under groupBy "file", or overall) and the marker-score behavior when a sequence is built. The main remaining gap is that a few tuning parameters (fast, waitSeconds, gapSeconds) have no explanation anywhere, which matters for a 10-parameter tool.

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 70%, so many parameters (keep, frame, sourceFolder, allowOriginal, groupBy) are already documented in the schema. The description adds real value for the conditional behavior of groupBy (per-file vs overall output shape) and sequenceName (builds a sequence with the score on each marker), but leaves fast, gapSeconds, and waitSeconds completely unexplained in both places.

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 ("Rank the takes of a footage folder") and immediately names the basis of ranking (audio cues the media engine heard), so an agent knows this is not a generic sort but a cue-driven take selection. It is clearly separable from editing siblings and from rough_cut_by_cues, with which it only shares a cache.

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 clear operating context: rank takes, optionally build a selects sequence via sequenceName, and "Always check the picks by eye" signals results are heuristic and need human/agent verification. It does not explicitly state when to prefer this over rough_cut_by_cues or find_shots, so routing is inferred rather than spelled out.

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