Perform a semantic scene

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

Annotations already declare this is a mutating, non-idempotent, non-destructive call. The description adds one genuinely useful behavioral disclosure: YouAndEye owns timing, transitions, scrolling completion, and neutral return, which tells the agent it should not try to time beats itself. It does not disclose what cancel does to an in-flight sequence, what status returns, or how interrupt_policy=reject behaves, all of which matter for a stateful performance tool.

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?

Two tight sentences with no filler, and the three action verbs are front-loaded. The second sentence is slightly cryptic ('neutral return' is undefined) but the overall shape is efficient and the reader gets the gist immediately.

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 7-parameter, zero-coverage, stateful mutation tool with nested beat objects and policies for waiting, interruption, and return, the description is far too thin. Output schema exists so return values need not be described, but the action-beats relationship, the wait/interrupt/return policy semantics, and the meaning of 'bounded' (maxItems 16) are all absent. The agent is left to reverse-engineer the call contract from the schema alone.

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% for 7 parameters with 3 enums, so the description carries the full burden. It touches only 'pacing words and semantic modifiers', which loosely maps to PerceptionBeat.pace and SemanticModifiers, but leaves action, beats, wait_for_completion, wait_timeout_ms, interrupt_policy, and return_policy completely unexplained in both schema and description. An agent cannot know that action=status takes no beats or that wait_timeout_ms bounds the blocking call.

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 names three verbs (start, inspect, cancel) matching the action enum and a resource ('bounded sequence of emotional beats'), which is clear conceptually. However, the tool name 'perform' and title 'Perform a semantic scene' are vague and the description does not differentiate it from siblings like express or neutral, leaving the agent to guess the boundary between a 'beat sequence' and a single expression.

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 description lists the possible actions but never states when to pick perform over express, neutral, or configure_profile. There is no exclusion guidance, no prerequisite note, and no hint that 'beats' is only relevant for action=start and ignored for status/cancel. The 'Supply pacing words...' half-sentence implies a start-only context but does not say it.

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