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

With annotations declaring readOnlyHint=false and idempotentHint=false, the description still adds genuine value: the 12 MB decoded input cap, the pricing model (charged by audio duration beyond a free allowance), and the video exclusion. It does not describe failure behavior on oversized input or what happens to partial results, which keeps it below 5.

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?

Three tight sentences, front-loaded with the operation and output, then constraints, then cost. No filler or restatement of the name.

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?

With no output schema, the description usefully names the return structure (segments + word timestamps) and covers the key cost and size constraints for a paid, open-world tool. It falls short on parameter-level meaning and error handling for the low-coverage 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?

Schema description coverage is only 20% (just request_id), so the description must compensate for glossary, language, mime_type, and audio_base64 — and it does not mention any of them. Only the size limit loosely touches audio_base64; glossary and language semantics are 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?

States a specific verb (transcribe), the resource (extracted audio), and the return shape (segments and word timestamps). This clearly separates it from siblings like translate_captions and correct_captions, which operate on already-produced captions.

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 the prerequisite that audio must already be extracted ("extracted audio", "Never send video") and a hard size ceiling, but it names no sibling alternative and gives no when-not-to-use guidance beyond the video exclusion. Usage is implied rather than explicitly routed.

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