character_speak

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

Annotations already declare non-readOnly, non-idempotent, non-destructive, closed-world. The description adds real operational context beyond that: playback occurs on physical hardware, the bridge de-duplicates playback when the text appears in the final answer, uncertain playback should not be retried, and text is capped at ~600 chars / 30 seconds. These are meaningful behavioral traits not derivable from the annotations.

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 action and device effect, followed by practical constraints and pitfalls in short sentences. Every sentence carries an instruction the agent needs; only minor tightening is possible, such as folding the duplicate-playback note into the 'do not also call speak' line.

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 non-idempotent hardware-mutating tool with no output schema, the description covers side effects, retry policy, and post-call bookkeeping well. Remaining gaps are secondary: `intensity` behavior and what happens when no character is configured or the display Pi is unreachable.

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%, so the description has to carry parameter meaning. It adds guidance for `emotion` (match the reply, not the user's feelings), explains that `voice=default` uses the user's saved preference, and constrains `text` to under 600 characters / 30 seconds. `intensity` is never mentioned and the non-default voice options are left entirely to the enum, so the compensation is substantial but not complete.

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 ('Speak as the configured character'), and adds the distinct device-side effects (matching face, animated mouth on the display Pi) that separate it from the plain `speak` sibling. An agent can distinguish it from `speak` and `character_expression` without opening any 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?

Explicit alternatives and exclusions are given: 'Do not also call speak for the same reply' and 'Choose emotion to match the reply, not infer the user's feelings.' It also covers post-call behavior ('put the spoken text in the final answer') and a failure policy ('Do not automatically retry uncertain playback'). This is close to a complete routing instruction.

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