live_start

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

With no annotations, the description carries the full burden and does well: it discloses background execution, optional monitor-page and OSC streaming, opt-in hands/pose cost, and a privacy guarantee (blendshapes and head pose, "no identities"). It omits failure modes such as a camera already in use or what happens on a second start.

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 action and source set, then the tracked signals, then the consent rule and next steps. Dense but nearly every clause adds usable information; the parenthetical affect and blendshape detail runs long.

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 needn't be spelled out, and the description covers sources, opt-ins, consent, and follow-up tools for a six-parameter tool. The unexplained mic parameter is the notable hole.

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 0%, so the description must compensate and largely does: source formats (camera, camera:N/name, screen:N, rtsp/srt/YouTube URL, file), watch_for with a concrete example, hands with pinch/gesture meaning, pose as body, and osc as output routing. The mic parameter and its "auto" default are never explained.

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 (start live perception in the background) and enumerates the accepted source types plus everything it tracks. It is clearly distinguished from live_now/live_snapshot/live_stop, which it names as follow-on tools, and from frame/watch style one-shot siblings.

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 an explicit consent constraint ("Only start the camera or microphone when the user asked for it") and routes to the three follow-up tools. It does not name an alternative for non-realtime or one-shot needs, so it stops short of full when-not guidance.

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