Image Foundry

Image Foundry

MCP server for generating and editing images via PackyAPI's gpt-image-2 model, enabling Codex to create and modify images based on prompts.

Category
Visit Server

README

Image Foundry

一个调用 PackyAPI gpt-image-2 的本地图片工作台,同时提供标准 MCP Server,方便 Codex 直接调用。

1. 配置 API Key

MCP Server 从环境变量或项目根目录的 .env 读取 Sora 分组令牌,不把令牌放进代码或 Codex 配置文件。最方便的方式是:

cp .env.example .env
# 然后编辑 .env,填入 PACKY_API_KEY

.env 已加入 .gitignore,不会被提交。

可选环境变量:

export PACKY_API_BASE_URL="https://www.packyapi.com"
export IMAGE_FOUNDRY_OUTPUT_DIR="$HOME/Code/create_image/output"

2. 接入 Codex

在终端执行:

codex mcp add image_foundry --env PACKY_API_KEY="$PACKY_API_KEY" --env PACKY_API_BASE_URL="https://www.packyapi.com" -- node "$HOME/Code/create_image/mcp-server.mjs"

确认注册成功:

codex mcp get image_foundry --json

如果你不想使用 .env,也可以让 MCP 进程继承当前 shell 的环境变量:

codex mcp add image_foundry -- node "$HOME/Code/create_image/mcp-server.mjs"

注册后重新打开一个 Codex task,Codex 就能发现这些工具:

  • generate_image:根据提示词生成图片
  • edit_image:编辑本地图片,可选传入 mask
  • get_image_config:查看模型、API 地址和输出目录,不会返回 API Key

示例请求:

请用 generate_image 生成一张横向 1536x1024 的图片:雨夜的上海街头,霓虹灯倒映在湿地面上,电影感摄影。

生成文件默认位于 ~/Code/create_image/output/。调用 edit_image 时,把本地图片的绝对路径传给 image_path。 编辑请求默认不发送 input_fidelity,行为与 PackyAPI 官方 curl 示例一致;确实需要最大程度保留原图细节时可显式传入 input_fidelity: "high"

3. 启动本地 Web 服务(可选, 可以不依赖智能体直接使用 Web 端制图)

进入当前项目后执行:

npm run dev

然后打开 http://127.0.0.1:4173/。页面只调用本地同源接口,API Key 由 Node 后端从项目根目录的 .env 读取,不会进入页面、浏览器存储或网络请求代码。

可用配置:

IMAGE_FOUNDRY_WEB_PORT=4173
PACKY_REQUEST_TIMEOUT_MS=600000

图片生成和高清编辑经常超过 60 秒,默认上游超时为 10 分钟。服务日志会输出请求 ID、接口阶段、尺寸、质量、文件大小、耗时和状态码,但不会记录 API Key、图片内容或完整提示词。

4. MCP 传输说明

mcp-server.mjs 使用 MCP 标准 stdio JSON-RPC 传输,不依赖第三方 npm 包,Codex 通过启动命令拉起它。图片编辑的上游上传使用 macOS/Linux 自带的 curl(powershell 可能需要额外适配),避免 Node fetch 在长时间图片生成连接中约 60 秒被中断;API Key 通过 curl 配置 stdin 传递,不会出现在进程命令行。服务只向 stdout 输出 MCP 消息,诊断信息写入 stderr,适合被 Codex 稳定托管。

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured