Stage Behaviour Run

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

With no annotations, the description does real work: it discloses the two documented failure modes (no behaviour on the object, unknown action), states that a failing function answers with `error` and emits behaviour_error, and names the events emitted (behaviour_run, behaviour_state on state change). It also clarifies that audio plays unless sound=False. It does not mention permissions, side-effect duration, or whether the call blocks until the function completes — the main remaining gaps for a state-mutating trigger.

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?

The core behaviour is front-loaded in the first clause and the supporting detail (sound flag, errors, reply shape, events) follows in a compact block with no filler sentences. It is dense and slightly run-on, with the reply-shape notation {object, action, via, state, error?} duplicated by the output schema, but nothing is padding.

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?

For a five-parameter state-changing trigger with zero annotation coverage this is reasonably complete — error paths, events, and silent-mode are all covered, and the output schema handles the return shape. The unexplained `scene` parameter and the unstated default for `action` (a caller omitting it silently presses) are the gaps that would most likely cause a mis-call.

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 carry all parameter meaning, and it covers four of five: object (must have a behaviour), action (the menu label), value (an input value), and sound (silent test). The required `scene` parameter is never explained, and no defaults are mentioned (action defaults to 'press', sound to true, value to null), leaving the agent to guess what a scene identifier looks like and what happens when action is omitted.

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?

The description states a concrete action — invoke an object's behaviour the way a VR user would, either by pressing a menu item (action=label) or by writing a value to one of its inputs — and names the page command it maps to (behaviour_run). It is distinguishable from siblings like stage_behaviours (listing) and stage_behaviour_state (state read) by the fact that it triggers execution and emits behaviour_run. The framing 'Do what the person does in VR' is metaphorical before it becomes literal, but the clarification arrives immediately.

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?

Usage is implied through two modes (action vs value) and one modifier (sound=False for a silent test), and the error note 'no such action (the reply lists them)' hints that discovery is possible via the failure reply. However, it never explicitly says to call stage_behaviours first to learn valid actions, nor when to prefer a value input over an action press. The agent must infer the sequencing from context.

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