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

Annotations only supply readOnlyHint=true, so the description carries the real burden and does: it discloses the price formula (list price x 1.4 + $0.0005, min $0.005), that the 402 quotes the exact charge, that settlement matches the quote, and that billing happens only on a returned image. These are material facts an agent could not infer otherwise. It omits failure modes beyond the 402 and says nothing about latency or content-policy rejection.

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 action and return type, then dense but purposeful constraint text with no filler. It is a single dense block with heavy parenthetical nesting, which makes the model/parameter mapping slower to parse than a short list would be.

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 7-parameter generation tool with 0% schema coverage and no output schema, the description supplies model IDs, per-family parameter applicability, defaults, cost mechanics, billing condition, input limit, and the response form (image block). Only output_format is left uncovered, which is a minor omission against everything else.

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 0%, so the description must compensate, and it largely does: it enumerates valid model IDs with a default, states which models accept size/quality vs resolution/aspect_ratio, and gives defaults for quality (medium) and aspect_ratio (1:1) plus allowed aspect ratios. The gap is output_format (png/jpeg/webp), which is documented in neither the schema nor the description, and the 4000-character prompt limit is stated but not tied to the model choice.

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 and resource plus the return shape: 'Generate one image from a prompt and return it as an image block.' That is far more than a restatement of the name. It does not reference the sole sibling 'models', so an agent gets no explicit routing cue between the two.

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 usable selection logic by splitting parameters along the model family axis (openai/* takes size+quality, google/* takes resolution+aspect_ratio) and names the default model and defaults. It stops short of saying when not to use this tool or how to pick a model for a given intent beyond naming them, and never mentions the 'models' sibling.

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