read which face belongs to each speaker

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

Annotations already establish readOnlyHint/openWorldHint=false, and the description goes well beyond them: it discloses the three status values, the revision field, the structure of matched vs uncertain speakers, the faceTrack URL format (public JSON, one box per frame), and the failure reason taxonomy. It also clarifies that wait waits on the analysis rather than the transcription, which is a subtle behavioral distinction an agent would otherwise get wrong.

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 scope, then the conditional routing, then the return shape, then the next step — a sensible order with no filler sentences. It is dense and long, and the slice-speaker-face-track guidance appears in both the return-value and next-step sentences, which is slight redundancy but each occurrence serves a different point.

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?

No output schema exists, yet the description fully describes the response envelope (status, revision, speakers with matched/uncertain branches and their fields), so an agent knows exactly what it will receive. Combined with the error-reason routing and the named downstream tool, nothing needed to call this correctly is missing.

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 both parameters are already documented in the schema, including the ~50s wait semantics and the 409 on unfinished transcription. The description adds real meaning beyond that by contrasting wait here with jobs_wait with kind speaker_analysis, making the wait target unambiguous; the id parameter itself gets no extra semantics beyond 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?

States a specific verb and resource: read which face belongs to each transcript speaker, by transcription id, for a COMPLETED transcription. It explicitly distinguishes itself from the two most confusable siblings, transcriptions_analyze-speakers (which starts/retries an analysis) and transcriptions_slice-speaker-face-track (the downstream consumer of the returned revision).

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 explicit when/when-not routing: normally the analysis runs automatically, so only call analyze-speakers when this returns unavailable with reason not_requested, analysis_failed or analysis_timeout, and it names the other unavailable reasons (retry_limit, source_binding_unavailable, source_too_long, too_many_speakers) that do NOT warrant a retry. It also routes uncertain speakers to speaker detect or lipsync: false, and points to jobs_wait for the same analysis.

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