scenes

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

With no annotations, the description carries the full burden, and it does disclose real behavior: scenes_to_markers writes markers, scenes_split actually cuts clips (a mutation), and the runtime dependency determines whether PySceneDetect or ffmpeg answers, with the result naming the backend. It stops short of saying whether the split is destructive to source files, whether it is reversible, or what permissions/backups are needed.

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 the resource, then packed with action semantics and defaults in a couple of dense sentences; nothing is padding. The telegraphic fragments ('ffmpeg without it — the result names which one answered') are terse but still parseable, at a slight cost to readability.

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?

There is no output schema, and the description compensates by stating that detect_scenes reports cuts in timeline time and that results name the answering backend. Combined with the three action descriptions and the arg defaults, an agent has enough to invoke it correctly; the main missing piece is the destructive/reversibility profile of scenes_split.

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?

The schema only exposes a generic 'args' object plus a bare action enum, so the description is doing the real work: it enumerates filepath, optional clip_name, backend with its auto|content|adaptive|ffmpeg values, threshold, and the min_scene_len=0.5 default. That is genuine added meaning over the schema, though it does not explain how threshold interacts with the backend choice.

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?

Opens with a precise resource statement ('Shot boundaries') and then names all three sub-actions with a specific verb+object for each: detect_scenes reads sources and reports cuts, scenes_to_markers writes a marker per cut, scenes_split cuts the clips. The three internal operations are mutually distinguished, and no sibling tool (inspect, edit, mark, transcript, etc.) overlaps this resource.

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 implies a workflow ordering — detect_scenes produces the cuts, the other two consume them — and notes which backend (PySceneDetect vs plain ffmpeg) will answer. However it never states when to prefer detect_scenes alone versus chaining to markers or split, nor any exclusion or prerequisite, so action selection is left to inference.

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