cw_ui_recipe

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

With no annotations, the description carries the full disclosure burden and does real work: it warns that synthetic keystrokes do not reach SONAR, that the menu bar is reachable via AXExpand, and that JUCE modals can block the app with a recovery path. These are non-obvious, empirically grounded constraints that materially change agent behavior. What it does not disclose is the shape of the response — whether it returns prose guidance, a checklist, or structured steps.

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 sentences, front-loaded with the capability and followed by the sharpest practical caveats (keystrokes, AXExpand, modals). The phrase 'Measured, not guessed' is a small piece of framing flavor rather than an operational fact, but it does signal that these claims are empirically verified, which is useful trust 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?

For a zero-parameter reference tool with no output schema and no annotations, the description covers scope, key gotchas, and recovery adequately. The remaining gap is that it never says what invoking it actually yields or how comprehensive the recipe is, so an agent cannot judge whether one call suffices or whether it should be called repeatedly as the UI state changes.

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?

The tool takes zero parameters and the schema is an empty object, so there is nothing for the description to disambiguate. Baseline 4 applies; the description correctly does not waste words on parameter discussion.

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 opens with a specific verb and subject — driving SONAR's own UI via a computer-use runtime — and scopes it to reachability plus JUCE modal-dialog recovery. That is clearly distinct from the command-execution siblings (cw_run_command, cw_navigate, cw_file_dialog), though it never names those alternatives explicitly. An agent can tell this is a knowledge/recipe tool rather than an action tool, but only by inference from the content.

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 rather than stated: you consult this when automating SONAR's GUI, especially when a modal blocks the app. There is no explicit 'use this instead of X' routing, no statement about when UI automation is the wrong choice (e.g. prefer cw_run_command when a command exists), and no note on when to re-read it. The 'how to recover when a JUCE modal dialog blocks the app' clause is the closest thing to a triggering condition.

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