Transcribe audio/video (word timestamps, SRT/VTT)

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

Annotations already declare readOnlyHint=false, idempotentHint=true, openWorldHint=true and destructiveHint=false, so the safety profile is covered. The description adds genuinely new behavior: the service is free, local files are uploaded on the caller's behalf, output is capped at 100 MB, and outputPath causes a file to be written to disk. Auth requirements and latency/queue behavior are the only notable omissions.

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 capability, then inputs, then outputs, then options — a logical order. It is dense and parenthetical-heavy in the return-value sentence, but essentially every clause carries information an agent needs; nothing is pure filler.

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 fully by describing the three return shapes (words[], cues[], transcript) and the SRT/VTT alternative with its granularity switch. Combined with the stated input types and size limit, an agent has everything needed to call this correctly on the first attempt.

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 60% schema coverage, the description carries real weight: it explains format=srt|vtt, granularity=cue|word (including the karaoke/alignment use case), and that outputPath only applies to subtitle formats. It also clarifies the semantic difference between raw `words[]` timing and legibility-stretched `cues[]`, which the schema does not convey. The language parameter is only restated as a list, adding little.

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 ("Speech-to-text") with the engine named (Deepgram Nova-3) and enumerates the accepted inputs (asset id, public URL, local file). It also distinguishes itself from siblings like vivid_analyze_audio or vivid_generate_voice by making clear this produces transcripts/subtitles rather than analysis or synthesized speech.

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 clear operating context: three source types, three languages, a 100 MB ceiling, and the format/granularity switches that select between JSON and subtitle files. It does not explicitly say when to prefer this over vivid_analyze_audio or when not to use it, so there are no stated exclusions or named alternatives.

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