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

Annotations already declare readOnly, idempotent, non-destructive and openWorld, so the safety profile is covered. The description adds genuinely new behavioral facts: the input is a fixed bundled image (so the call takes no user data) and results are cached (so repeated calls are cheap and return identical output). It does not describe the response shape, but that is minor here.

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?

One short sentence with zero waste; the key differentiators (free, demo, fixed sample, cached) are all front-loaded and none of the words are filler.

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 parameterless, cached demo tool with no output schema, the definition covers what it operates on and that the input is fixed and cached. The only minor gap is that it never states the output is extracted OCR text, which a reader must infer from the tool name.

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?

With zero parameters, the baseline is 4. The description reinforces this by stating the input is a fixed sample PNG, making it clear an agent must not attempt to pass an image path or URL to this tool — a useful implicit contract for a demo endpoint.

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 names the parent tool (image_to_text_ocr) and specifies the demo runs on a fixed sample invoice PNG, which immediately separates it from the real OCR tool and the other *_demo siblings. It stops short of a crisp verb phrase like 'run OCR on a bundled sample image', but the resource and mode are unambiguous.

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?

'Free demo' and 'fixed sample' imply this is for trying the capability without supplying your own image, and the sibling name points to the production tool. However, there is no explicit statement such as 'use when you have no image input; call image_to_text_ocr to process your own image', so the routing decision is left to inference.

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