OpenAI-Compatible Images plugin
Generate, edit and deliver images through your API providers or the Codex ChatGPT route, with a canvas and version history.
- 63 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Stores provider keys in local configuration and sends requests to external services.
Install
- Works in
Manual install
codex plugin marketplace add Syh1906/openai-compatible-imagegenThe first README command adds the marketplace; then `codex plugin add openai-compatible-imagegen@openai-compatible-imagegen`.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Stores provider keys in local configuration and sends requests to external services.
This is third-party code. Review the repository files before installing.
What it does
OpenAI-Compatible Images gives agents one image core in two installation shapes. A standalone skill suits Codex CLI, Claude Code and OpenCode, while the Codex App plugin adds result cards, a canvas for marking regions, immutable artifacts and version history. API key profiles support the OpenAI-compatible, Atlas, xAI Images and Gemini protocols. Providers, aliases and parameters are configured separately and a model is chosen per request. There are JSONL batches, transparency handling, quality checks and long asynchronous jobs whose results can be recovered after a wait times out.
Who it is for. Designers and developers who edit images in Codex or Claude Code and keep their own keys for several providers.
Good fit when
- An article cover is generated and refined by marking regions on a canvas.
- A single request picks a model among several connected providers.
- A batch of dozens of prompts is processed from a JSONL file.
- A result is delivered with a transparent background and checked first.
Not a fit when
- There are no API keys and the ChatGPT route is unavailable in your Codex version.
- You need native operation in an environment without Python 3.12.
- Canvas editing of animated GIFs and SVGs is required, which is not supported.
Example request
Generate a 16:9 cover about query optimization, open the canvas and fix the marked area.Limitations
Python 3.12 or newer and an image service key are required; the plugin also needs Git, Node.js 20 and a Codex version with plugin support. The plugin and the standalone skill are installed separately with different local settings and artifact folders, and a migration guide covers the move. The canvas and local transforms do not work for GIF and SVG. The ChatGPT route depends on the Codex App itself. Image requests go to the chosen provider under your key and are billed there.
How to disable. Remove the plugin through Codex or the skill folder from the skills directory. Delete saved artifacts and the key configuration separately.
FAQ
How does the plugin differ from the skill?
The skill is portable and works in terminal clients. The Codex App plugin adds result cards, a canvas, artifacts and versions. Install one per environment.
Which providers work?
API key profiles cover the OpenAI-compatible, Atlas, xAI Images and Gemini protocols, including two Gemini flavors. MuAPI, for example, connects through the OpenAI-compatible route.
What if waiting for generation is cut off?
Jobs are asynchronous and long-running: confirmed results can be recovered and local processing resumed from saved originals. See the long-running generation guide.
Is a restart needed after installing?
Yes, quit Codex completely and start it again once so it loads the skill, MCP tools and bundled dependencies.
Can GIF and SVG be edited?
They can be requested from supporting APIs, previewed and exported. The canvas and local raster transforms are unavailable for them.
Discussion
No questions or reviews yet.
Related
PluginsImage generation and editing from any agent: the imagegen-agent CLI and a skill with the Codex prompt library for compatible APIs
A skill and CLI for image and video generation through Gemini, OpenAI and Kie AI, with provider choice and AI-content marking
The media-pipeline plugin generates images through Gemini or OpenAI gpt-image-2: a skill, CLI and MCP server install in one step.
A plugin that draws and edits images in Claude Code through the built-in image_gen of local Codex on a ChatGPT login, no API key.
Claude Image (GPT Image 2)
claude-image
A skill teaching agents to work with GPT Image 2. OpenAI API access is needed and it is unavailable from Russia without workarounds
Skill, MCP server and CLI that let any agent call Codex image generation, after picking a style from several directions
README in short
One image core ships as a skill and a plugin. The docs cover model configuration, long-running jobs, migration and updating.