screen_to_text

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

With no annotations provided, the description carries full responsibility and delivers thoroughly. It discloses that output is from Tesseract OCR (best-effort, not a UI tree), lists specific limitations (cannot report widget types, enabled/checked state, focus, scroll position, off-screen content, misreads small/low-contrast text), and warns that the 'words' detail can cost more tokens than a screenshot. This goes well beyond the minimum.

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 appropriately sized given the tool's complexity. It front-loads the core purpose in the first sentence, then efficiently covers cost motivation, parameter details, limitations, and alternatives in a well-structured flow. Every paragraph earns its place, and there is no fluff or redundancy.

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?

Given no annotations or output schema, the description covers purpose, usage, limitations, return format (reading order text with bounding boxes), and parameter semantics for most parameters. The only gap is min_confidence, which is left unspecified. Despite this, the description is largely complete for an agent to use the tool safely and effectively.

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 schema has no descriptions (coverage 0%), so the description must compensate. It does for two of three parameters: detail (explains 'lines', 'text', 'words' with trade-offs) and image_path (explains omission captures the monitor). However, min_confidence is never mentioned, so the agent is left guessing what it controls. This slight omission prevents a perfect score.

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 clear, specific action: 'Read a screenshot as compact structured text instead of sending an image.' It specifies the resource (screenshot/image) and the output (structured text with bounding boxes). It also distinguishes itself from sibling tools like ocr_image and ocr_current_monitor by framing it as a token-saving alternative and contrasting with Playwright's accessibility tree.

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?

Explicitly states when to use this tool: for surfaces without an accessibility tree (native desktop apps, remote desktops, canvas/WebGL, video frames, scanned documents). It gives a clear alternative: 'If the target is a web page, use Playwright's accessibility tree instead' and explains why. It also provides operational guidance on omitting image_path to capture the current monitor and how to choose the detail level.

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