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

The annotations only declare that this writes, is open-world and non-idempotent; the description goes far beyond them with the price discovery order (platform reads the price before spending, returns estimatedCredits), the charge/refund lifecycle, and the three terminal states including the dangerous "unknown" case where the connection dropped and a charge may have occurred. The no-job-id caveat and the double-charge warning on resubmission are exactly the behavioral facts an agent cannot get from structured fields.

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?

It is front-loaded with purpose, then cost, then failure modes, then the resubmission warning, which is the right order for an agent scanning for go/no-go signals. The pricing parenthetical and refund detail are dense but each clause carries an operational consequence; the single longest sentence could be split for readability.

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?

There is no output schema, and the description compensates by naming the returned fields (estimatedCredits, state with its "failed"/"unknown"/"running" values, refunded) and where artifacts land (the gallery as audio). For a charged, non-idempotent generation call, an agent has everything needed to estimate cost, avoid duplicate charges, and recover from an ambiguous result.

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 100%, so baseline is 3, but the description adds real meaning the schema lacks: the byte-based billing that makes accented and non-Latin text cost up to 3x, worked cost examples (200 chars = 2 credits, 1,000 = 5), and the default voice. The only unstressed parameter is `wait`, whose 3-minute versus ~45 s tradeoff is left entirely to the schema.

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?

"Turns text into a spoken voiceover MP3" states a specific verb, resource and output format, then enumerates concrete use cases (narration, ad read, podcast intro, character line). It also names the sibling that resolves voice questions (aetherwave_list_voices), so the agent can separate this from generate_music and the voice-listing tool without opening a schema.

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?

It gives explicit routing rules for the non-obvious cases: scripts over 1000 characters must be split at sentence ends into one call per part, voice lookup goes to aetherwave_list_voices, and a result with state "running" must be retrieved via aetherwave_list_my_creations type "audio" rather than resubmitted. It also names a guard parameter ("pass maxCredits to refuse anything dearer"), leaving little to inference.

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