openai-image

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

With no annotations at all, the description carries the full behavioral burden and does disclose the key trait that it writes files to disk and returns their paths. However, for a 19-parameter mutation tool it says nothing about overwrite behavior for existing filenames, required credentials, latency/cost, or failure modes — meaningful gaps when no annotations cover the mutation profile.

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 sentences, both front-loaded with the operation and then the output behavior. There is no filler, no restatement of parameters, and no redundancy with the schema.

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?

It correctly covers the return semantics ('returns their paths') in the absence of an output schema, which is important given 19 parameters and file output. But the description is thin for a tool this complex: it omits file-overwrite behavior, credential prerequisites, and the relationship to the generate sibling, leaving real gaps for an agent preparing a call.

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 100%, so the schema already documents nearly all 19 parameters and a baseline of 3 would be defensible. The description still adds value beyond the schema by stating the 'up to 16 reference images' cap (the schema only specifies minItems: 1 with no maximum) and by framing mask/markup as optional guidance mechanisms.

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 names a specific verb and resource ('Edit images or compose from up to 16 reference images'), identifies the underlying API (images.edit), and adds scoping detail (mask or markup overlay, files saved to output_dir). It is clear what the tool does, though it never explicitly contrasts itself with the sibling openai_image_generate beyond the implicit edit-vs-generate distinction.

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?

Usage is only implied: the agent can infer this is for transforming existing images rather than creating new ones. There is no explicit statement of when to choose this over openai_image_generate, nor any guidance on preconditions (e.g. images must exist locally) beyond what the schema states.

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

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

With no annotations, the description does disclose the key behavioral fact that this writes files to output_dir and returns their paths, which tells the agent this is a mutating, persistent operation. It omits other relevant traits: cost/latency of image generation, whether existing files are overwritten, and what happens on partial failure.

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 sentences, front-loaded with the action and followed by the persistence/return behavior; nothing is padded. The parenthetical API name is slightly redundant but harmless.

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 16-parameter tool with nested objects, no annotations, and no output schema, the description covers the essentials (what it produces, where files go, what comes back) but leaves the agent without guidance on format/size tradeoffs, cost, or the edit-vs-generate decision. The schema carries most of the remaining detail, so it is adequate rather than 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 description coverage is 100%, and the schema already explains formats, sizes, frame counts, model names, and the SVG sub-object in detail. The description adds only the cross-parameter link between GIF output and output_format=gif, so the baseline of 3 is appropriate.

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 ("Generate images (OpenAI images.generate)") and enumerates the three output kinds (still, animated GIF, SVG), plus the side effect of saving files. It distinguishes itself naturally from the sibling openai_image_edit by being the generation path, though it never names that sibling explicitly.

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?

Usage is only implied: an agent can infer this is for creating new images rather than editing existing ones, but the description never says "use openai_image_edit to modify an existing image" or states any exclusion or precondition beyond the SVG-access note carried in the schema. No guidance on when to prefer one model or format.

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