glass_a11y_marks

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

Annotations already declare readOnlyHint=true, so no safety contradiction. The description adds valuable behavioral context beyond annotations: boxes 'can drift' and the user should 'inspect action results', plus the error condition 'without a tree'. This informs the agent about reliability and failure modes.

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?

Three tightly packed sentences: the core action, the intended usage of results, and a caveat plus alternative. Every sentence earns its place, and the most critical information is front-loaded.

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, read-only tool with no output schema, the description tells the agent what the tool produces, how to consume the result, what can go wrong, and which sibling to fall back to. Nothing essential for correct invocation is missing.

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 has zero parameters, so schema coverage is trivially 100% and the baseline is 4. The description adds no parameter syntax, but none is needed; it instead clarifies that the produced IDs are output artifacts, not inputs, which is relevant for post-processing.

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 resource: 'Capture a screenshot with numbered accessibility boxes and a text legend.' This clearly distinguishes it from siblings like glass_screenshot (plain screenshot) and glass_a11y_snapshot (accessibility tree), and it adds the distinctive detail that the boxes carry numbers usable with click_element.

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 explicit when-to-use and when-not-to-use guidance: 'Use the IDs with a click_element action' names the downstream interaction pattern, and 'Errors without a tree; use glass_screenshot' explicitly tells the agent to fall back to another tool when the accessibility tree is unavailable.

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