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

Annotations already declare readOnly, idempotent, non-destructive behavior, so the bar is lower, and the description goes well beyond them: experimental GPU availability, a $0.50 USD per-call price, no free preview, x402 paid execution, and the fact that an unpaid MCP call returns payment instructions rather than results. That is exactly the kind of cost-and-failure-mode context an agent needs before invoking. The only gap is that it does not characterize output beyond pointing at the output schema.

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 purpose is front-loaded and each early sentence is short, but the middle section is padded with redundant boilerplate and the duplicated, grammatically broken 'Selection:' clause is pure noise. The pricing/payment sentences earn their place; the redundant when-clause repeated in the selection line does not.

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?

An output schema exists, so return values need not be explained, and the payment/availability disclosure covers the main operational risk. What's missing for a two-required-parameter image tool is any guidance on the base64 payload (encoding expectations, size ceiling) or on supported-versus-rejected mime types, leaving the agent to reverse-engineer the schema.

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%, so the description carries the burden, yet it only says 'Input is a strict JSON object; see the schema' and adds no meaning. It never mentions that imageBase64 must be base64-encoded or that the schema caps it at ~2.8MB (maxLength 2796204), which is a practical constraint an agent could easily violate. It also doesn't explain mimeType's role, though the enum itself is self-explanatory in the schema.

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?

States a specific verb-plus-resource ('Extract text and document layout from a single image') and names the underlying model (Baidu Unlimited-OCR), which is concrete enough for an agent to know what it does. However, it never distinguishes itself from plausible siblings such as document_facts_extract, and the 'Selection:' line degenerates into a circular restatement ('prefer for Use when an agent needs to extract text from a PNG, JPEG, or WebP image.'), so differentiation is not actually achieved.

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 only when-to-use guidance is 'Use when an agent needs to extract text from a PNG, JPEG, or WebP image', which merely restates the purpose and adds no decision criteria. There are no when-not conditions, no named alternatives (e.g. document_facts_extract or preview_capability), and the 'supporting; prefer for…' clause is malformed and conveys no routing information.

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