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

Annotations only supply readOnlyHint=false; everything else comes from the description, which discloses the $0.07 fixed price, that images are stored in this workspace rather than provider links, that no-download_url results explain themselves, and which failures are refunded (SIMPLER_IMAGE_OUTPUT_FLAGGED, SIMPLER_EDIT_RESULT_UNDELIVERED, PROMPT_CHECK_UNAVAILABLE 'nothing charged'). The enumerated refusal codes tell the agent exactly how to recover instead of retrying blindly.

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 cost, purpose and the mask rule, then output shape, then failures, then NEXT — a sensible order. The error-code run is long and dense, but each entry carries a distinct recovery action, so little is truly wasted.

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 fully specifies the return payload ({images:[{url, asset_key, download_url?, width, height}], price, next}) plus preview behavior, and covers the failure surface and chaining path. An agent can call this and handle outcomes without consulting anything 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 100%, so the baseline is 3; the description still adds meaning by restating the mask convention ('white = background'), noting an empty prompt triggers VALIDATION_ERROR ('describe the scene'), and explaining the asset_key handoff that ties mask_asset_key/image_asset_key to prior results. It does not restate request_ref idempotency, which the schema already covers in depth.

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 ('Replace a photo's background with a generated scene') and pins the mechanism (FLUX Fill) plus the required mask semantics ('white = background'). That distinguishes it from siblings like image.cut_background (which removes rather than generates), image.inpaint and image.replace_sky without needing to open a schema.

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 operational context: you must pass a mask covering the background, and it states the follow-on path ('pass images[0].asset_key as image_asset_key to the next image.* tool'). It never explicitly says when to choose this over image.inpaint or image.cut_background, so exclusions are missing.

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