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

Goes well past the annotations (which only say it's a non-destructive, closed-world write): it discloses the credit cost formula (1/second + 1, 11 credits for 10s), that HD costs more, the multi-minute async latency, the exact polling tool and identifier to use, and a refund guarantee for failures. Those are the four things an agent most needs to know before spending a user's credits on a long-running job.

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 purpose, then cost, then the pre-condition, then the async protocol – a sensible priority order with no filler sentences. The cost sentence is slightly dense ("1 credit a second plus 1 (11 credits for 10 seconds)") but every clause carries operational information.

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 paid, asynchronous, non-destructive generation tool with a full output schema, the description covers everything an agent needs: what it produces, what it costs, what to agree before calling, how long it takes, how to check status, and what happens on failure. Nothing material is left to inference.

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 100% and the schema already documents hd, part, prompt, songId and seconds including ranges and defaults, so the structured data carries most of the burden. The description adds real economic meaning to `seconds` and `hd` (cost per second, HD surcharge) and implicitly links "the scene" to `prompt`, but adds no new semantics for songId or part beyond the schema. Baseline 3 is right.

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+resource ("Make a music video for one of their songs") and immediately qualifies the output: a described scene animated to a portion of the track with the song underneath. This is unmistakably distinct from the other generation siblings (generate_song, generate_remix, reimagine_song, change_lyrics), which produce audio rather than video.

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 pre-call guidance ("Agree the scene and length with them first") and a concrete follow-up path ("poll get_video_details with the videoId"), which is exactly what an agent needs to sequence this call. It stops short of stating explicit exclusions or naming which sibling to use if the user only wants audio, but the usage context is unambiguous.

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