creator-factory-mcp
Allows generating images through the OpenAI Images API by explicitly selecting the openai provider, with jobs tracked and assets retrieved through the MCP server.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@creator-factory-mcpsubmit two images of a blue cube with the mock provider"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Creator Factory Core
A local-first Python image task engine: SQLite queue, independent worker, deterministic Mock Provider, optional OpenAI Images Provider, local asset manifests, CLI, and STDIO MCP. It has no website, remote server, message broker, MySQL, or dependency on an existing private AI Creator application.
中文简介:这是一个可独立运行的通用图片任务核心。默认只使用 Python、SQLite 和本地文件;Mock Provider 可在完全离线的 CI 中生成确定性的合法 PNG。真实 OpenAI 请求必须显式选择 openai Provider,并通过环境变量提供密钥。
Quick start
python -m pip install -e ".[mcp]"
creator-factory init
creator-factory submit "a fictional blue cube" --provider mock --count 2
creator-factory-worker --once
creator-factory status JOB_ID
creator-factory assets --job-id JOB_IDRuntime files live under .creator-factory/ by default and are ignored by Git.
Related MCP server: openai-image-gen
OpenAI Images Provider
The optional provider uses the official Image API generation endpoint, defaults to gpt-image-2, and reads only OPENAI_API_KEY from the process environment:
set OPENAI_API_KEY=...
creator-factory submit "a fictional editorial illustration" --provider openai
creator-factory-worker --onceThe key is never accepted in CLI/MCP arguments, stored in SQLite, or written to logs. CI never sends a real request. A real-provider job is marked before network execution; if its Worker is interrupted, the job becomes interrupted_no_retry and requires inspection/manual resubmission to avoid accidental duplicate billing.
MCP tools
creator_submit_imagecreator_job_statuscreator_job_cancelcreator_job_retrycreator_assets_listcreator_asset_get
Submission is asynchronous and returns creator.job.v1; asset rows use creator.asset_manifest.v1. Run creator-factory-mcp as a local STDIO server and keep creator-factory-worker as a separate process.
Safety
Prompts and outputs stay local unless the openai provider is selected. Do not submit confidential prompts or third-party personal data without authorization. Generated outputs still require human review for rights, accuracy, and policy compliance.
MIT licensed.
This server cannot be deployed
Maintenance
Related MCP Connectors
Create and manage AI image and video generations through Quriov's fixed public MCP tools.
Generate AI images and videos from any compatible MCP client.
Generate images with any major model — one API key, one prepaid balance, one MCP.
Focused MCP server for OpenAI image/audio generation (v2.0.0). Wraps endpoints via HAPI CLI.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceProvides tools for generating and editing images using OpenAI's gpt-image-1 model via an MCP interface, enabling AI assistants to create and modify images based on text prompts.15Apache 2.0
- AlicenseNot gradedqualityCmaintenanceExposes multiple image generation backends as independent MCP tools for generating images with configurable models and parameters.MIT
- AlicenseAqualityBmaintenanceGenerate images with OpenAI and Google Gemini directly from any MCP client, saving assets into your project with batch support and explicit file output.22 npmMIT
- AlicenseAqualityCmaintenanceEnables LLMs to generate images via MCP by calling AI models from providers like SiliconFlow, OpenAI, or custom APIs, with tools for image generation and model listing.2MIT