export_training_data

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 the concrete artifacts each layout emits (tar, shards + manifest, train.jsonl + images), which is real value. However it says nothing about whether `out` is created or overwritten, disk cost, or that exports can run for hours — significant gaps for a heavy mutation-style operation.

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 layout list are front-loaded in the first sentence, and the closing sentence about sources/height/dpi carries real information. The parenthetical framework annotations on each layout add slight bulk but plausibly help layout selection.

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 10-parameter tool with no annotations and no output schema, the description covers the most decision-relevant axis (format) well but omits the semantics of half the parameters and any statement about output location handling or failure modes. Adequate to start an invocation, not complete enough to invoke confidently.

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 0%, so the description must compensate, and it does add crucial meaning for the `format` values (the schema has no enum, so the five layout names are documented only here) plus the non-obvious `height` (ladder stage in lines) and `dpi` (at reference width). But it leaves `out`, `ffmpeg_dir`, `instruction`, `timeout_seconds`, `max_samples_per_shard`, and `reference_width_inches` entirely undocumented.

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?

Names a specific verb (Export) and resource (Moviola frames as a training dataset) and enumerates the five output layouts it can produce, which lets an agent confirm capability without opening the schema. This is clearly distinct from siblings like probe_media, convert_media, and training_backends.

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 usage by describing acceptable sources ('any media files or folders') and output layouts, but never states when to pick this over convert_media or training_backends, nor any prerequisites like requiring ffmpeg. Usage is inferable but not explicit.

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