Rotate a local video without re-encoding

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

Annotations already declare not-read-only, not-destructive, not-open-world; the description adds substantial behavior beyond that: stream-copy without decoding, full packet/timeline/matrix verification, hard refusals for fragmented or edited timelines, an AAC-only audio rule, and the important caveat that pixels are not baked so player support varies. It stops short of saying what happens if output_path already exists (overwrite vs. fail), which is the main remaining gap for a mutation tool.

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 core action and tightly packed — nearly every sentence carries a distinct constraint or caveat. It is telegraphic rather than verbose, though 'Requires no browser' is a low-value trailing note that adds little for an agent choosing between tools.

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?

For a mutation tool with no output schema, the description covers inputs, codec/container preconditions, refusals, and a caveat about downstream rendering. It is close to complete, missing only overwrite/conflict behavior for output_path and any hint about the return payload.

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?

With 50% schema description coverage, the description compensates for the undocumented parameters: 'Sound must be AAC or explicitly omitted' explains the constraint governing keep_audio, and 'display matrix rotation' clarifies that degrees produces a metadata rotation rather than re-encoded pixels. The path parameters and the 90/180/270 enumeration are only repeated from the schema, not enriched.

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 ('Copy an admitted MP4/MOV video into a new MP4') plus the exact operation applied (90/180/270 display matrix rotation). This distinguishes it from sibling video tools like abox_video_trim, abox_video_join and abox_video_remux, which do not apply rotation.

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 gives acceptance preconditions ('standard orientation and unsquashed geometry only', 'fragmented or edited timelines are refused', 'Sound must be AAC or explicitly omitted'), which implicitly tells the agent when the tool will fail. However, it never names an alternative path when those preconditions don't hold, nor when to prefer this over remux/re-encode tools; the agent must infer that.

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