Generate voice (TTS)

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

Annotations only say it is a non-read-only, non-destructive, non-idempotent, closed-world call; the description goes well beyond that by disclosing per-provider credit costs (1 / 1 / 0.5 / free), that output is a public MP3 URL kept only 7 days, and that saved voices require account-holder ownership, biometric consent and 18+ confirmation in the web app. Those are exactly the constraints an agent needs before spending credits or promising persistence, and nothing contradicts the annotations.

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 the anti-cloning exclusion, then structured provider-by-provider with costs inline. It is dense and slightly run-on in the provider sentence, but nearly every clause carries actionable information, so little is wasted.

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 an 8-parameter generation tool with no output schema, it covers the return value (public MP3 URL, 7-day expiry, optional local download), cost model, and the consent boundary around saved voices. Missing pieces are secondary input constraints — text length limit, speed range, locale default — which an agent would only learn by reading the schema.

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 schema coverage at 38%, the description carries most of the load, and it does so for the hardest parameters: it defines what `voice` means per provider (preset name, preset id or saved-voice id, model id, English description) and confirms `emotion` is minimax-only. It leaves `speed`, `locale`, `filename` and the 3000-character text cap entirely to an undocumented schema, which is the remaining gap.

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 ("Synthesize speech from text") plus the accepted voice categories (preset, designed, saved). It explicitly disambiguates from a plausible sibling capability by saying it never clones from an audio file or URL, so an agent can tell what this tool is and is not.

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 routing guidance per provider: use vivid_list_voices for MiniMax ids, omnivoice for description-based design, deepgram for free Aura-2, gemini as the cheap default. It also rules out cloning. It stops short of naming alternatives for non-TTS needs (e.g. transcribe, generate_music), but within the TTS decision space the guidance is explicit.

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