check_testability

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

No annotations are provided, so the description carries the full burden of behavioral disclosure. The verb 'Audit' strongly implies a read-only operation, and the mention of 'Compose testTag fix' suggests it provides advice rather than applying changes. However, the description does not explicitly state that it never modifies anything, nor does it describe the output or return value. It also doesn't clarify any prerequisites (e.g., a connected device) or side effects. Given the absence of annotations, this is a moderate gap; the description gives some context but not full transparency.

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 description is a single, well-structured sentence that front-loads the core purpose and then details the counts and the Compose fix. It includes the platform caveat and documentation reference without fluff. Every clause earns its place: the main action, the specific metrics, and the platform-specific note. It is concise (~50 words) and immediately usable, earning a top score.

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?

The tool is relatively simple (one optional parameter, no output schema), but the description omits important context: it does not describe the output format (does it return a report? JSON? a count?), the meaning of 'device_id', or any prerequisites like a connected device or active screen. It does provide platform guidance, but the missing parameter semantics and return-value information leave the description incomplete for a thorough understanding. Given the simplicity, it could easily have been more complete, so a 3 is appropriate.

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%, meaning the input schema provides no descriptions for the single parameter 'device_id'. The tool description does not mention 'device_id' at all, so it adds no meaning beyond the schema's bare name. Since the description must compensate for low schema coverage, its failure to explain what 'device_id' refers to (e.g., an Android device identifier) leaves the agent guessing. This is a significant gap for even one parameter, so a score of 2 is warranted.

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 specific verb ('Audit'), a clear resource ('the current screen'), and a specific goal ('locator stability'). It breaks down exactly what it does: counts how many actionable elements have a stable id, which don't, and mentions the Compose testTag fix. This clearly distinguishes it from all sibling tools, which are actions (tap, swipe, etc.) or queries (dump_ui, find_element). No ambiguity about its purpose.

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 gives clear platform context: 'Android; on iOS, id-coverage is not meaningful — see docs/ios-testing.md'. This serves as an explicit exclusion for iOS and directs to documentation. However, it does not explicitly compare this tool to any alternative tools (like dump_ui or describe_element) or state when to use it vs. them. It implies usage (when you need to audit stability) but lacks explicit 'use this instead of X when...' guidance. Overall, it provides solid context without enumerating alternatives, earning a 4.

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