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

Annotations only carry readOnly=false / idempotent=false / destructive=false; the description adds the costly operational detail an agent actually needs: ~1.5 credits per minute billed by the second, no double charge for the same range, in-place replacement of existing words within the range, and that a null/null whole-file call replaces a transcript built from ranges. The billing and mutation-overwrite behavior is not inferable from any structured field.

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 purpose and sized reasonably for the number of behaviors covered, but it is delivered as one dense run-on block and clauses like 'such a call answers that transcript and runs nothing' and 'Bind with transcript= in film props' are compressed to the point of requiring a second read.

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?

Covers cost, caching, overwrite semantics and downstream binding, which is thorough for a 4-param tool with no output schema. It does not say whether the call blocks or starts a job (wait_for_jobs appears only in the schema's projectId text), which is a real gap for a long-running transcription operation.

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 75% and both time parameters are already documented in the schema, so the baseline would be 3. The description goes further by explaining the range's behavioral meaning (only that range is read, its words replace prior words in that range) and the null/null whole-file semantics, which the schema only partially conveys.

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 ('Transcribe a ready video or sound') plus the output form ('word-timed speech, in source seconds'), which immediately distinguishes it from read_transcript, patch_transcript and split_sound in the sibling set. An agent knows exactly what this does without opening the 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?

Explicitly prescribes the precondition ('Search assets first'), the when-not case ('if transcript is completed for the whole file, do not call transcribe again... such a call answers that transcript and runs nothing'), the alternative path (pass startTime/endTime for just the shot range), and the repair case (re-transcribing a range to fix a wrong or deleted word). Nothing is left to inference.

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