Stage Take Transcribe

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

With no annotations at all, the description carries the full burden and does real work: it discloses the write side effect (creates voice.json), the alignment semantics ('words on the take's clock, snapped onto the measured voice'), and a hard runtime prerequisite ('Needs the speech server (speakwright) running'). What is missing is failure behavior when that server is down and whether an existing voice.json is overwritten.

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?

One front-loaded sentence opens with the action and resource, then layers the condition, the side effect, and the prerequisite — nothing wasted and no preamble. The nested parenthetical makes the middle slightly dense, but every clause carries information.

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?

An output schema exists, so return values need not be described, and the description covers side effects and the external dependency that would otherwise cause silent failure. The remaining gap is the undocumented 'scene' parameter, which is material for a tool that cannot be invoked without it.

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 0% and both parameters are bare strings, so the description must compensate. It effectively documents 'take' via the takes/<id>/voice.json path, but 'scene' is never mentioned or explained — an agent gets no clue whether it is a scene id, name, or path. Partial compensation only.

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 take's own audio') plus the exact artifact produced ('writes takes/<id>/voice.json'), which is far more concrete than the name alone. It implicitly separates itself from the automatic transcription pipeline ('takes recorded before takes were transcribed as they land') but never names a sibling tool, so an agent must infer the boundary from context.

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?

The parenthetical gives clear triggers: takes predating automatic transcription, or a take whose transcription failed. That is a genuine when-to-use condition rather than an implied one. However, no when-not condition and no named alternative (e.g. any re-transcription or sync sibling) is offered, so the routing is one-sided.

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