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

Goes well beyond the hints: it discloses the 500-read price threshold with the confirm=true gate, that video output is asynchronous and takes minutes, that re-asking the same prompt returns the pending/ready file and is charged once, that cost is metered by what the file cost to make, that results land in the user's Puley history, and that sexual/violent/deceptive content is rejected. Those are exactly the operational facts an agent needs before spending the user's credits.

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 one-line purpose, then grouped by kind, then by cross-cutting controls and pricing. It is dense and delivered as a single run-on paragraph, so bullet or line-break structure would aid scanning, but nearly every clause carries information specific to a parameter or behavior.

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 13-parameter generation tool with no schema descriptions and no output schema, it covers mode selection, per-mode options, cost/confirm gating, latency, result location, and content policy — essentially everything needed to call it correctly and set expectations. The only soft spot is a mild tension with idempotentHint=false, since re-asking is described as returning the same job charged once.

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 description coverage is 0%, so the description carries the full burden and does: all 13 parameters are explained, including enumerated values for voice (narrator, british woman, comedian, etc.), emotion, quality (quick/standard/best), shape (square/portrait/story/landscape or a ratio like 4:5), a speed range of 0.5–2, named model options, seconds as video length, and the image vs image_link distinction.

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 specific verb+resource ('Make a file from a description') and then enumerates the four distinct modes (image, video, voice, music) it produces. This clearly separates it from the neighbouring research/data siblings (find_videos, watch_video, trend_to_video) which retrieve rather than generate.

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 concrete conditions that select each mode: image for a picture or an edit when image/image_link is supplied, video for a clip or an animated photo, voice for words read by a named voice, music for a ~30s instrumental. It also states the confirm=true precondition for jobs over 500 reads and that model may be left empty for a good low-cost default. It stops short of naming alternatives (e.g. list_models for browsing models, trend_to_video for trend-driven clips).

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