can1357/oh-my-pi · error · Error

No image API credentials found. Connect a Codex (ChatGPT)…

Error message

No image API credentials found. Connect a Codex (ChatGPT) subscription, use a GPT Responses/Codex model with OpenAI credentials, log in with google-antigravity or xAI Grok OAuth, or set OPENAI_API_KEY, XAI_API_KEY, OPENROUTER_API_KEY, GEMINI_API_KEY, GOOGLE_API_KEY, or DEEPINFRA_API_KEY.

What it means

Before attempting any provider, the image-gen tool probes for credentials across all supported image providers (OpenAI/Codex, google-antigravity, xAI OAuth, OpenAI, xAI, OpenRouter, Gemini/Google, DeepInfra keys). If none are found it throws this error instead of attempting requests. It aggregates every supported auth path into one actionable message.

Solutions

Set at least one supported key env var, e.g. export OPENAI_API_KEY=sk-... (or GEMINI_API_KEY / OPENROUTER_API_KEY / XAI_API_KEY / DEEPINFRA_API_KEY).
Or run /login and connect xAI Grok OAuth or google-antigravity.
Or use a Codex (ChatGPT) subscription / GPT Responses model with existing OpenAI credentials.
Confirm the env var names exactly (GOOGLE_API_KEY vs GOOGLE_API_TOKEN etc.) and that the process can see them.
If keys live in a .env file, ensure the CLI loads it or export them in the invoking shell.

Example fix

// before (no credentials)
$ omp -p "draw a cat"
// after
$ export GEMINI_API_KEY=AIza...
$ omp -p "draw a cat"

Defensive patterns

Strategy: validation

Validate before calling

const IMAGE_KEY_VARS = ['OPENAI_API_KEY','XAI_API_KEY','OPENROUTER_API_KEY','GEMINI_API_KEY','GOOGLE_API_KEY','DEEPINFRA_API_KEY'] as const;
const hasImageCreds = IMAGE_KEY_VARS.some(v => process.env[v]) || oauthConnected('xai') || oauthConnected('antigravity');
if (!hasImageCreds) throw new Error('Configure at least one image provider credential before calling image generation');

Try / catch

try {
  await imageGen(params);
} catch (err) {
  if (err instanceof Error && err.message.startsWith('No image API credentials found')) {
    printSetupInstructions(); // point user to /login or env var export
  } else throw err;
}

Prevention

Check credential presence at CLI startup and show setup guidance early
Keep env vars in a sourced .env so non-interactive runs see them
Prefer one documented primary provider plus its key in CI secrets
Use exact variable names from the error message

When it happens

Trigger: Invoking the image generation tool in an environment with no OPENAI_API_KEY, XAI_API_KEY, OPENROUTER_API_KEY, GEMINI_API_KEY, GOOGLE_API_KEY, DEEPINFRA_API_KEY, no Codex (ChatGPT) subscription connection, and no google-antigravity or xAI OAuth login, at image-gen.ts:1751.

Common situations: Fresh install with no /login performed; CI runners without secrets exported; dotenv/.env not loaded by the CLI; keys present but under wrong variable names; shell profile not sourced in non-interactive contexts.

Understand the failure class

Background: "API key is required" / "API key not found" / "No API key was set": the missing-api-key error family across 16 libraries — this error's family across 16 libraries.

Related errors

AI-assisted analysis of can1357/oh-my-pi@9690622007 (2026-08-31). Data as JSON: /api/errors/586f952662dc71a9. Report an issue: GitHub.

Appendix: source

Thrown at packages/coding-agent/src/tools/image-gen.ts:1751

							model,
							imageCount: inlineImages.length,
							imagePaths,
							images: inlineImages,
							responseText,
							promptFeedback: data.promptFeedback,
							usage: data.usageMetadata,
						},
					};
				} catch (error) {
					if (!(error instanceof ProviderHttpError) || requestSignal?.aborted) {
						throw error;
					}
					failures.push({ provider, error });
				}
			}

			if (!foundCredentials) {
				throw new Error(
					"No image API credentials found. Connect a Codex (ChatGPT) subscription, use a GPT Responses/Codex model with OpenAI credentials, log in with google-antigravity or xAI Grok OAuth, or set OPENAI_API_KEY, XAI_API_KEY, OPENROUTER_API_KEY, GEMINI_API_KEY, GOOGLE_API_KEY, or DEEPINFRA_API_KEY.",
				);
			}

			if (failures.length === 0 && unsupportedAspectRatioProvider) {
				assertImageAspectRatioSupported(unsupportedAspectRatioProvider, params.aspect_ratio);
			}

			if (failures.length === 0 && editUnsupportedProvider) {
				throw new Error(
					`${editUnsupportedProvider} image generation is text-to-image only and cannot edit input images. Configure an edit-capable provider (openai, openai-codex, antigravity, xai, openrouter, gemini) or retry without input images.`,
				);
			}

			throw new AggregateError(
				failures.map(failure => failure.error),
				`Image generation failed for all credentialed providers: ${failures.map(failure => failure.provider).join(", ")}`,
			);

View on GitHub (pinned to 9690622007)