vultr/openclaw
1
0
Fork 0
mirror of https://github.com/openclaw/openclaw.git synced 2026-05-06 09:40:43 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: support transparent OpenAI image generation

Browse Source
This commit is contained in:
Peter Steinberger
2026-04-25 19:28:25 +01:00
parent 0bf4876add
commit de0097a23c
9 changed files with 362 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
CHANGELOG.md
View File

@@ -87,6 +87,7 @@ Docs: https://docs.openclaw.ai
- Gateway/dashboard: render Control UI and WebSocket links with `https://`/`wss://` when `gateway.tls.enabled=true`, including `openclaw gateway status`. Fixes #71494. (#71499) Thanks @deepkilo.
- Agents/OpenAI-compatible: default proxy/local completions tool requests to `tool_choice: "auto"` when tools are present, so providers enter native tool-calling mode instead of replying with plain-text tool directives. (#71472) Thanks @Speed-maker.
- OpenAI image generation: use `gpt-5.5` for the Codex OAuth responses transport instead of the retired `gpt-5.4` model, fixing 500s from ChatGPT Codex image generation. Fixes #71513. Thanks @baolongl.
- OpenAI image generation: route transparent-background default-model requests to `gpt-image-1.5`, document the expected `image_generate` call shape, and keep Azure/custom OpenAI-compatible deployment names untouched. Thanks @steipete.
- Google video generation: download direct MLDev Veo `video.uri` results instead of passing them through the Files API path, fixing 404s after successful generation/polling. Fixes #71200. Thanks @panhaishan.
- Google video generation: fall back to the REST `predictLongRunning` Veo endpoint for text-only SDK 404s while keeping reference image/video generation on the SDK path. Fixes #62309 and #63008. (#62343) Thanks @leoleedev.
- MiniMax music generation: switch the bundled default model from the unsupported `music-2.5+` id to the current `music-2.6` API model. Fixes #64870 and addresses the music default from #62315. Thanks @noahclanman and @edwardzheng1.

4
docs/gateway/config-agents.md
View File

@@ -342,8 +342,8 @@ Time format in system prompt. Default: `auto` (OS preference).
- Also used as fallback routing when the selected/default model cannot accept image input.
- `imageGenerationModel`: accepts either a string (`"provider/model"`) or an object (`{ primary, fallbacks }`).
- Used by the shared image-generation capability and any future tool/plugin surface that generates images.
- Typical values: `google/gemini-3.1-flash-image-preview` for native Gemini image generation, `fal/fal-ai/flux/dev` for fal, or `openai/gpt-image-2` for OpenAI Images.
- If you select a provider/model directly, configure matching provider auth too (for example `GEMINI_API_KEY` or `GOOGLE_API_KEY` for `google/*`, `OPENAI_API_KEY` or OpenAI Codex OAuth for `openai/gpt-image-2`, `FAL_KEY` for `fal/*`).
- Typical values: `google/gemini-3.1-flash-image-preview` for native Gemini image generation, `fal/fal-ai/flux/dev` for fal, `openai/gpt-image-2` for OpenAI Images, or `openai/gpt-image-1.5` for transparent-background OpenAI PNG/WebP output.
- If you select a provider/model directly, configure matching provider auth too (for example `GEMINI_API_KEY` or `GOOGLE_API_KEY` for `google/*`, `OPENAI_API_KEY` or OpenAI Codex OAuth for `openai/gpt-image-2` / `openai/gpt-image-1.5`, `FAL_KEY` for `fal/*`).
- If omitted, `image_generate` can still infer an auth-backed provider default. It tries the current default provider first, then the remaining registered image-generation providers in provider-id order.
- `musicGenerationModel`: accepts either a string (`"provider/model"`) or an object (`{ primary, fallbacks }`).
- Used by the shared music-generation capability and the built-in `music_generate` tool.

20
docs/providers/openai.md
View File

@@ -27,6 +27,7 @@ changing config.
| GPT-5.5 with ChatGPT/Codex subscription auth | `openai-codex/gpt-5.5` | Default PI route for Codex OAuth. Best first choice for subscription setups. |
| GPT-5.5 with native Codex app-server behavior | `openai/gpt-5.5` plus `embeddedHarness.runtime: "codex"` | Forces the Codex app-server harness for that model ref. |
| Image generation or editing | `openai/gpt-image-2` | Works with either `OPENAI_API_KEY` or OpenAI Codex OAuth. |
| Transparent-background images | `openai/gpt-image-1.5` | Use `outputFormat=png` or `webp` and `openai.background=transparent`. |
<Note>
GPT-5.5 is available through both direct OpenAI Platform API-key access and
@@ -254,8 +255,17 @@ See [Image Generation](/tools/image-generation) for shared tool parameters, prov
</Note>
`gpt-image-2` is the default for both OpenAI text-to-image generation and image
editing. `gpt-image-1` remains usable as an explicit model override, but new
OpenAI image workflows should use `openai/gpt-image-2`.
editing. `gpt-image-1.5`, `gpt-image-1`, and `gpt-image-1-mini` remain usable as
explicit model overrides. Use `openai/gpt-image-1.5` for transparent-background
PNG/WebP output; the current `gpt-image-2` API rejects
`background: "transparent"`.
For a transparent-background request, agents should call `image_generate` with
`model: "openai/gpt-image-1.5"`, `outputFormat: "png"` or `"webp"`, and
`openai.background: "transparent"`. OpenClaw also protects the public OpenAI and
OpenAI Codex OAuth routes by rewriting default `openai/gpt-image-2` transparent
requests to `gpt-image-1.5`; Azure and custom OpenAI-compatible endpoints keep
their configured deployment/model names.
For Codex OAuth installs, keep the same `openai/gpt-image-2` ref. When an
`openai-codex` OAuth profile is configured, OpenClaw resolves that stored OAuth
@@ -275,6 +285,12 @@ Generate:
/tool image_generate model=openai/gpt-image-2 prompt="A polished launch poster for OpenClaw on macOS" size=3840x2160 count=1
```
Generate a transparent PNG:
```
/tool image_generate model=openai/gpt-image-1.5 prompt="A simple red circle sticker on a transparent background" outputFormat=png openai='{"background":"transparent"}'
```
Edit:
```

58
docs/tools/image-generation.md
View File

@@ -48,13 +48,14 @@ The agent calls `image_generate` automatically. No tool allow-listing needed —
## Common routes
| Goal | Model ref | Auth |
| ---------------------------------------------------- | -------------------------------------------------- | ------------------------------------ |
| OpenAI image generation with API billing | `openai/gpt-image-2` | `OPENAI_API_KEY` |
| OpenAI image generation with Codex subscription auth | `openai/gpt-image-2` | OpenAI Codex OAuth |
| OpenRouter image generation | `openrouter/google/gemini-3.1-flash-image-preview` | `OPENROUTER_API_KEY` |
| LiteLLM image generation | `litellm/gpt-image-2` | `LITELLM_API_KEY` |
| Google Gemini image generation | `google/gemini-3.1-flash-image-preview` | `GEMINI_API_KEY` or `GOOGLE_API_KEY` |
| Goal | Model ref | Auth |
| ---------------------------------------------------- | -------------------------------------------------- | -------------------------------------- |
| OpenAI image generation with API billing | `openai/gpt-image-2` | `OPENAI_API_KEY` |
| OpenAI image generation with Codex subscription auth | `openai/gpt-image-2` | OpenAI Codex OAuth |
| OpenAI transparent-background PNG/WebP | `openai/gpt-image-1.5` | `OPENAI_API_KEY` or OpenAI Codex OAuth |
| OpenRouter image generation | `openrouter/google/gemini-3.1-flash-image-preview` | `OPENROUTER_API_KEY` |
| LiteLLM image generation | `litellm/gpt-image-2` | `LITELLM_API_KEY` |
| Google Gemini image generation | `google/gemini-3.1-flash-image-preview` | `GEMINI_API_KEY` or `GOOGLE_API_KEY` |
The same `image_generate` tool handles text-to-image and reference-image
editing. Use `image` for one reference or `images` for multiple references.
@@ -93,7 +94,8 @@ Use `"list"` to inspect available providers and models at runtime.
</ParamField>
<ParamField path="model" type="string">
Provider/model override, e.g. `openai/gpt-image-2`.
Provider/model override, e.g. `openai/gpt-image-2`; use
`openai/gpt-image-1.5` for transparent OpenAI backgrounds.
</ParamField>
<ParamField path="image" type="string">
@@ -233,9 +235,10 @@ through the Codex Responses backend. Legacy Codex base URLs such as
`https://chatgpt.com/backend-api/codex` for image requests. It does not
silently fall back to `OPENAI_API_KEY` for that request. To force direct OpenAI
Images API routing, configure `models.providers.openai` explicitly with an API
key, custom base URL, or Azure endpoint. The older
`openai/gpt-image-1` model can still be selected explicitly, but new OpenAI
image-generation and image-editing requests should use `gpt-image-2`.
key, custom base URL, or Azure endpoint. The `openai/gpt-image-1.5`,
`openai/gpt-image-1`, and `openai/gpt-image-1-mini` models can still be
selected explicitly. Use `gpt-image-1.5` for transparent-background PNG/WebP
output; the current `gpt-image-2` API rejects `background: "transparent"`.
`gpt-image-2` supports both text-to-image generation and reference-image
editing through the same `image_generate` tool. OpenClaw forwards `prompt`,
@@ -260,8 +263,31 @@ OpenAI-specific options live under the `openai` object:
```
`openai.background` accepts `transparent`, `opaque`, or `auto`; transparent
outputs require `outputFormat` `png` or `webp`. `openai.outputCompression`
applies to JPEG/WebP outputs.
outputs require `outputFormat` `png` or `webp` and a transparency-capable OpenAI
image model. OpenClaw routes default `gpt-image-2` transparent-background
requests to `gpt-image-1.5`. `openai.outputCompression` applies to JPEG/WebP
outputs.
When asking an agent for a transparent-background OpenAI image, the expected
tool call is:
```json
{
"model": "openai/gpt-image-1.5",
"prompt": "A simple red circle sticker on a transparent background",
"outputFormat": "png",
"openai": {
"background": "transparent"
}
}
```
The explicit `openai/gpt-image-1.5` model keeps the request portable across
tool summaries and harnesses. If the agent instead uses the default
`openai/gpt-image-2` with `openai.background: "transparent"` on the public
OpenAI or OpenAI Codex OAuth route, OpenClaw rewrites the provider request to
`gpt-image-1.5`. Azure and custom OpenAI-compatible endpoints keep their
configured deployment/model names.
Generate one 4K landscape image:
@@ -269,6 +295,12 @@ Generate one 4K landscape image:
/tool image_generate action=generate model=openai/gpt-image-2 prompt="A clean editorial poster for OpenClaw image generation" size=3840x2160 count=1
```
Generate a transparent PNG:
```
/tool image_generate action=generate model=openai/gpt-image-1.5 prompt="A simple red circle sticker on a transparent background" outputFormat=png openai='{"background":"transparent"}'
```
Generate two square images:
```

152
extensions/openai/image-generation-provider.test.ts
View File

@@ -194,7 +194,12 @@ describe("openai image generation provider", () => {
const provider = buildOpenAIImageGenerationProvider();
expect(provider.defaultModel).toBe("gpt-image-2");
expect(provider.models).toEqual(["gpt-image-2"]);
expect(provider.models).toEqual([
"gpt-image-2",
"gpt-image-1.5",
"gpt-image-1",
"gpt-image-1-mini",
]);
expect(provider.capabilities.geometry?.sizes).toEqual(
expect.arrayContaining(["2048x2048", "3840x2160", "2160x3840"]),
);
@@ -428,6 +433,74 @@ describe("openai image generation provider", () => {
});
});
it("routes transparent default-model requests to the OpenAI image model that supports alpha", async () => {
mockGeneratedPngResponse();
const provider = buildOpenAIImageGenerationProvider();
const result = await provider.generateImage({
provider: "openai",
model: "gpt-image-2",
prompt: "Transparent sticker",
cfg: {},
outputFormat: "png",
providerOptions: {
openai: {
background: "transparent",
},
},
});
expect(postJsonRequestMock).toHaveBeenCalledWith(
expect.objectContaining({
url: "https://api.openai.com/v1/images/generations",
body: expect.objectContaining({
model: "gpt-image-1.5",
output_format: "png",
background: "transparent",
}),
}),
);
expect(result.model).toBe("gpt-image-1.5");
});
it("does not reroute transparent requests for custom OpenAI-compatible endpoints", async () => {
mockGeneratedPngResponse();
const provider = buildOpenAIImageGenerationProvider();
await provider.generateImage({
provider: "openai",
model: "gpt-image-2",
prompt: "Transparent custom endpoint sticker",
cfg: {
models: {
providers: {
openai: {
baseUrl: "https://openai-compatible.example.com/v1",
models: [],
},
},
},
},
outputFormat: "png",
providerOptions: {
openai: {
background: "transparent",
},
},
});
expect(postJsonRequestMock).toHaveBeenCalledWith(
expect.objectContaining({
url: "https://openai-compatible.example.com/v1/images/generations",
body: expect.objectContaining({
model: "gpt-image-2",
output_format: "png",
background: "transparent",
}),
}),
);
});
it("allows loopback image requests for the synthetic mock-openai provider", async () => {
mockGeneratedPngResponse();
@@ -684,6 +757,43 @@ describe("openai image generation provider", () => {
});
});
it("routes transparent default-model Codex OAuth requests to the alpha-capable image model", async () => {
mockCodexAuthOnly();
mockCodexImageStream({ imageData: "codex-transparent-image" });
const provider = buildOpenAIImageGenerationProvider();
const result = await provider.generateImage({
provider: "openai",
model: "gpt-image-2",
prompt: "Draw a transparent Codex sticker",
cfg: {},
authStore: { version: 1, profiles: {} },
outputFormat: "png",
providerOptions: {
openai: {
background: "transparent",
},
},
});
expect(postJsonRequestMock).toHaveBeenCalledWith(
expect.objectContaining({
url: "https://chatgpt.com/backend-api/codex/responses",
body: expect.objectContaining({
tools: [
expect.objectContaining({
type: "image_generation",
model: "gpt-image-1.5",
output_format: "png",
background: "transparent",
}),
],
}),
}),
);
expect(result.model).toBe("gpt-image-1.5");
});
it("uses configured Codex OAuth directly instead of probing an available OpenAI API key", async () => {
resolveApiKeyForProviderMock.mockImplementation(async (params?: { provider?: string }) => {
if (params?.provider === "openai") {
@@ -1213,6 +1323,46 @@ describe("openai image generation provider", () => {
);
});
it("does not reroute transparent background requests for Azure deployment names", async () => {
mockGeneratedPngResponse();
const provider = buildOpenAIImageGenerationProvider();
await provider.generateImage({
provider: "openai",
model: "gpt-image-2",
prompt: "Transparent Azure sticker",
cfg: {
models: {
providers: {
openai: {
baseUrl: "https://myresource.openai.azure.com",
models: [],
},
},
},
},
outputFormat: "png",
providerOptions: {
openai: {
background: "transparent",
},
},
});
expect(postJsonRequestMock).toHaveBeenCalledWith(
expect.objectContaining({
url: "https://myresource.openai.azure.com/openai/deployments/gpt-image-2/images/generations?api-version=2024-12-01-preview",
body: {
prompt: "Transparent Azure sticker",
n: 1,
size: "1024x1024",
output_format: "png",
background: "transparent",
},
}),
);
});
it("uses api-key header and deployment-scoped URL for .cognitiveservices.azure.com hosts", async () => {
mockGeneratedPngResponse();

36
extensions/openai/image-generation-provider.ts
View File

@@ -30,6 +30,7 @@ const DEFAULT_OPENAI_IMAGE_BASE_URL = "https://api.openai.com/v1";
const DEFAULT_OPENAI_CODEX_IMAGE_BASE_URL = OPENAI_CODEX_RESPONSES_BASE_URL;
const DEFAULT_OPENAI_CODEX_IMAGE_RESPONSES_MODEL = "gpt-5.5";
const OPENAI_CODEX_IMAGE_INSTRUCTIONS = "You are an image generation assistant.";
const OPENAI_TRANSPARENT_BACKGROUND_IMAGE_MODEL = "gpt-image-1.5";
const DEFAULT_OPENAI_IMAGE_TIMEOUT_MS = 180_000;
const DEFAULT_OUTPUT_MIME = "image/png";
const DEFAULT_OUTPUT_EXTENSION = "png";
@@ -52,6 +53,12 @@ const LOG_VALUE_MAX_CHARS = 256;
const MOCK_OPENAI_PROVIDER_ID = "mock-openai";
const OPENAI_OUTPUT_FORMATS = ["png", "jpeg", "webp"] as const;
const OPENAI_QUALITIES = ["low", "medium", "high", "auto"] as const;
const OPENAI_IMAGE_MODELS = [
DEFAULT_OPENAI_IMAGE_MODEL,
OPENAI_TRANSPARENT_BACKGROUND_IMAGE_MODEL,
"gpt-image-1",
"gpt-image-1-mini",
] as const;
const log = createSubsystemLogger("image-generation/openai");
const AZURE_HOSTNAME_SUFFIXES = [
@@ -186,6 +193,21 @@ function appendOpenAIImageOptions(
}
}
function resolveOpenAIImageRequestModel(
req: Parameters<ImageGenerationProvider["generateImage"]>[0],
options?: { allowTransparentDefaultReroute?: boolean },
): string {
const model = req.model || DEFAULT_OPENAI_IMAGE_MODEL;
if (
options?.allowTransparentDefaultReroute === true &&
model === DEFAULT_OPENAI_IMAGE_MODEL &&
req.providerOptions?.openai?.background === "transparent"
) {
return OPENAI_TRANSPARENT_BACKGROUND_IMAGE_MODEL;
}
return model;
}
function shouldAllowPrivateImageEndpoint(req: {
provider: string;
cfg: OpenClawConfig | undefined;
@@ -468,7 +490,7 @@ function createOpenAIImageGenerationProviderBase(params: {
id: params.id,
label: params.label,
defaultModel: DEFAULT_OPENAI_IMAGE_MODEL,
models: [DEFAULT_OPENAI_IMAGE_MODEL],
models: [...OPENAI_IMAGE_MODELS],
isConfigured: params.isConfigured,
capabilities: {
generate: {
@@ -517,7 +539,9 @@ function logCodexImageAuthSelected(params: {
authMode?: unknown;
timeoutMs: number;
}) {
const model = params.req.model || DEFAULT_OPENAI_IMAGE_MODEL;
const model = resolveOpenAIImageRequestModel(params.req, {
allowTransparentDefaultReroute: true,
});
log.info(
`image auth selected: provider=openai-codex mode=${sanitizeLogValue(
params.authMode,
@@ -549,7 +573,9 @@ async function generateOpenAICodexImage(params: {
transport: "http",
});
const model = req.model || DEFAULT_OPENAI_IMAGE_MODEL;
const model = resolveOpenAIImageRequestModel(req, {
allowTransparentDefaultReroute: true,
});
const count = resolveOpenAIImageCount(req.count);
const size = req.size ?? DEFAULT_SIZE;
const timeoutMs = resolveOpenAIImageTimeoutMs(req.timeoutMs);
@@ -711,7 +737,9 @@ export function buildOpenAIImageGenerationProvider(): ImageGenerationProvider {
transport: "http",
});
const model = req.model || DEFAULT_OPENAI_IMAGE_MODEL;
const model = resolveOpenAIImageRequestModel(req, {
allowTransparentDefaultReroute: publicOpenAIBaseUrl,
});
const count = resolveOpenAIImageCount(req.count);
const size = req.size ?? DEFAULT_SIZE;
const timeoutMs = resolveOpenAIImageTimeoutMs(req.timeoutMs);

39
src/agents/openai-ws-message-conversion.test.ts
View File

@@ -1,8 +1,45 @@
import { describe, expect, it } from "vitest";
import type { ResponseObject } from "./openai-ws-connection.js";
import { buildAssistantMessageFromResponse } from "./openai-ws-message-conversion.js";
import { buildAssistantMessageFromResponse, convertTools } from "./openai-ws-message-conversion.js";
describe("openai ws message conversion", () => {
it("preserves image_generate transparent-background guidance in OpenAI tool payloads", () => {
const [tool] = convertTools([
{
name: "image_generate",
description:
'Generate images. For transparent OpenAI backgrounds, use outputFormat="png" or "webp" and openai.background="transparent"; OpenClaw routes the default OpenAI image model to gpt-image-1.5 for that mode.',
parameters: {
type: "object",
properties: {
model: {
type: "string",
description:
"Optional provider/model override; use openai/gpt-image-1.5 for transparent OpenAI backgrounds.",
},
outputFormat: { type: "string", enum: ["png", "jpeg", "webp"] },
openai: {
type: "object",
properties: {
background: {
type: "string",
enum: ["transparent", "opaque", "auto"],
description:
"For transparent output use outputFormat png or webp; OpenClaw routes the default OpenAI image model to gpt-image-1.5 for this mode.",
},
},
},
},
},
},
]);
expect(tool?.description).toContain('openai.background="transparent"');
expect(tool?.description).toContain("gpt-image-1.5");
expect(JSON.stringify(tool?.parameters)).toContain("openai/gpt-image-1.5");
expect(JSON.stringify(tool?.parameters)).toContain("transparent");
});
it("preserves cached token usage from responses usage details", () => {
const response: ResponseObject = {
id: "resp_123",

68
src/agents/tools/image-generate-tool.test.ts
View File

@@ -218,6 +218,18 @@ describe("createImageGenerateTool", () => {
expect(createImageGenerateTool({ config: {} })).toBeNull();
});
it("tells agents how to request transparent OpenAI backgrounds", () => {
vi.stubEnv("OPENAI_API_KEY", "openai-key");
stubImageGenerationProviders();
const tool = requireImageGenerateTool(createImageGenerateTool({ config: {} }));
expect(tool.description).toContain('outputFormat="png" or "webp"');
expect(tool.description).toContain('openai.background="transparent"');
expect(tool.description).toContain("gpt-image-1.5");
expect(JSON.stringify(tool.parameters)).toContain("openai/gpt-image-1.5");
});
it("matches image-generation providers across canonical provider aliases", () => {
vi.spyOn(imageGenerationRuntime, "listRuntimeImageGenerationProviders").mockReturnValue([
{
@@ -595,6 +607,62 @@ describe("createImageGenerateTool", () => {
});
});
it("forwards transparent OpenAI background requests with a PNG output format", async () => {
const generateImage = vi.spyOn(imageGenerationRuntime, "generateImage").mockResolvedValue({
provider: "openai",
model: "gpt-image-1.5",
attempts: [],
ignoredOverrides: [],
images: [
{
buffer: Buffer.from("png-out"),
mimeType: "image/png",
fileName: "transparent.png",
},
],
});
vi.spyOn(mediaStore, "saveMediaBuffer").mockResolvedValue({
path: "/tmp/transparent.png",
id: "transparent.png",
size: 7,
contentType: "image/png",
});
const tool = createToolWithPrimaryImageModel("openai/gpt-image-1.5");
const result = await tool.execute("call-openai-transparent", {
prompt: "A transparent badge",
outputFormat: "png",
openai: {
background: "transparent",
},
});
expect(generateImage).toHaveBeenCalledWith(
expect.objectContaining({
cfg: expect.objectContaining({
agents: expect.objectContaining({
defaults: expect.objectContaining({
imageGenerationModel: { primary: "openai/gpt-image-1.5" },
}),
}),
}),
outputFormat: "png",
providerOptions: {
openai: {
background: "transparent",
},
},
}),
);
expect(result).toMatchObject({
details: {
provider: "openai",
model: "gpt-image-1.5",
outputFormat: "png",
},
});
});
it("includes MEDIA paths in content text so follow-up replies use the real saved file", async () => {
vi.spyOn(imageGenerationRuntime, "listRuntimeImageGenerationProviders").mockReturnValue([
{

10
src/agents/tools/image-generate-tool.ts
View File

@@ -96,7 +96,10 @@ const ImageGenerateToolSchema = Type.Object({
}),
),
model: Type.Optional(
Type.String({ description: "Optional provider/model override, e.g. openai/gpt-image-2." }),
Type.String({
description:
"Optional provider/model override, e.g. openai/gpt-image-2; use openai/gpt-image-1.5 for transparent OpenAI backgrounds.",
}),
),
filename: Type.Optional(
Type.String({
@@ -131,7 +134,8 @@ const ImageGenerateToolSchema = Type.Object({
openai: Type.Optional(
Type.Object({
background: optionalStringEnum(SUPPORTED_OPENAI_BACKGROUNDS, {
description: "OpenAI-only background hint: transparent, opaque, or auto.",
description:
"OpenAI-only background hint: transparent, opaque, or auto. For transparent output use outputFormat png or webp; OpenClaw routes the default OpenAI image model to gpt-image-1.5 for this mode.",
}),
moderation: optionalStringEnum(SUPPORTED_OPENAI_MODERATIONS, {
description: "OpenAI-only moderation hint: low or auto.",
@@ -570,7 +574,7 @@ export function createImageGenerateTool(options?: {
label: "Image Generation",
name: "image_generate",
description:
'Generate new images or edit reference images with the configured or inferred image-generation model. Set agents.defaults.imageGenerationModel.primary to pick a provider/model. Providers declare their own auth/readiness; use action="list" to inspect registered providers, models, readiness, and auth hints. Generated images are delivered automatically from the tool result as MEDIA paths.',
'Generate new images or edit reference images with the configured or inferred image-generation model. For transparent OpenAI backgrounds, use outputFormat="png" or "webp" and openai.background="transparent"; OpenClaw routes the default OpenAI image model to gpt-image-1.5 for that mode. Set agents.defaults.imageGenerationModel.primary to pick a provider/model. Providers declare their own auth/readiness; use action="list" to inspect registered providers, models, readiness, and auth hints. Generated images are delivered automatically from the tool result as MEDIA paths.',
parameters: ImageGenerateToolSchema,
execute: async (_toolCallId, args) => {
const params = args as Record<string, unknown>;