image_qa

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

Adds significant behavioral context beyond annotations: explains forwarding via MCP sampling, that OpenChrome never uses its own API keys, and details the unsupported response. No contradiction with annotations (readOnlyHint, destructiveHint, idempotentHint are all consistent).

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?

Two compact sentences plus a critical note. Information is front-loaded: purpose first, then mechanism, then constraints. Every sentence adds value with no 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?

Despite having no output schema, the description covers purpose, mechanism, dependency on client capability, fallback behavior, input requirements, and defaults. For a moderately complex tool (vision QA, three input modes, nested object), this is complete.

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 baseline is 3. The description adds meaning by stating 'REQUIRED Exactly one of...' for the screenshot object, clarifying that mime_type defaults to image/png, and explaining max_tokens defaults to 512. This goes beyond the schema descriptions.

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 clearly states the tool asks the host LLM a question about a screenshot. It uses a specific verb ('Ask') and resource ('screenshot'), and distinguishes from sibling tools like vision_find by specifying the mechanism (MCP sampling).

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?

Describes when the tool works (client must advertise sampling capability) and what happens otherwise (returns unsupported status). It also specifies that exactly one screenshot identifier must be supplied. However, it does not explicitly compare to alternatives like vision_find or state when not to use this tool.

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