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

Annotations already cover safety (readOnlyHint, destructiveHint=false, openWorldHint, idempotentHint=false), and the description adds genuinely new operational context: $0.01/call, payment via x402 USDC or a card key, and failed calls are free. 'Reads or computes only and changes nothing' largely restates the annotations, which caps this below 5.

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?

Front-loaded with the core capability and question mode, then constraints and cost. It is a bit repetitive, since the accepted formats and 3 MB cap duplicate the schema and 'changes nothing' duplicates the readOnly annotation, but every sentence otherwise carries useful information.

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?

With no output schema, the description still tells the agent what it gets back (a detailed description including visible text), plus input limits, pricing, and that failures are free. Only the exact return shape (plain text vs. structured) and latency are left unspecified.

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 coverage is 100%, so the schema already documents url and question fully, including formats and the 300-character question limit. The description restates the format/size constraint and explains the two modes, adding context but no syntax beyond the schema, matching the baseline 3.

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+resource: a vision model that returns a detailed description of an image at a URL, or answers a question about it. It also names the concrete artifacts it handles (visible text, screenshots, charts), which separates it from the ai_* text siblings and from transcribe.

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?

Gives clear when-to-use cases: alt text, OCR-style screenshot reading, chart reading, content checks, plus the two modes (default description vs. pass a question). It stops short of explicitly naming an alternative or exclusion (e.g., when to prefer transcribe or image), so no full when-not guidance.

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