
Flux-2 Online MCP
io.github.rocnubie由 GitHub 账号 rocnubie 发布v0.1.0更新于 Oct 24, 2026
Flux-2 Online is an online interface for the Flux 2 image generation model with prompt tuning and
概览
让助手通过只读工具和资源查询 Flux-2 Online 的图像生成风格、价格、常见问题与官方链接。
- 可以做什么
- 向 MCP 客户端提供 Flux-2 Online 的知识面。工具包括 list_styles(返回站点提供的图像生成风格与预设)、get_pricing(返回规范的价格入口)和 get_official_links(返回官网、支持与文档链接)。同时提供风格、价格、FAQ 和链接等资源,以及概述站点和推荐起始风格的提示。所有输出均为 text/markdown,且为只读。
- 适用场景
- 当用户向助手询问 Flux-2 Online、其图像生成风格、模型档位、价格或官方链接,或希望针对某个目标获得起始风格建议时适用。它是参考与发现类助手,并不用于实际生成或编辑图片。
- 运行要求
- 作为本地 stdio 进程在用户机器上运行,仅支持桌面客户端。可通过 Smithery CLI 安装,或用 git、pnpm 和 Node.js 从源码安装,然后在 MCP 客户端配置中指向 src/index.mjs。未声明需要 API 密钥、账号、环境变量或请求头。
安装
在 SourceWeft 中
- 打开 控制台中的 Flux-2 Online MCP,将其添加到工作区。
- 为需要使用其工具的对话启用该服务。
Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。
其他 MCP 客户端
参照 仓库 中的启动说明。
README
Flux-2 Online MCP Server
Flux-2 Online - Flux 2 AI Image Generator
[MCP Badge] [License: MIT] [Node] [smithery] [MCP]
A Model Context Protocol server that exposes the canonical Flux-2 Online knowledge surface — image generation workflows and styles, pricing, FAQ, official links — to MCP-compatible AI clients such as Claude Desktop, Cursor, Windsurf, and Continue. Read-only, no API keys, no quota, ~50 ms cold start.
Official website: https://flux-2.online
🎨 About Flux-2 Online
Flux-2 Online is a web-based AI image and video generation platform built around the Flux family of models. Users describe what they want in a text prompt, choose a model suited to their needs — from the fast and budget-friendly Flux 2 Dev to the higher-fidelity Flux 2 Pro or the flexible Flux 2 Flex — and receive generated images within seconds. Beyond text-to-image, the platform bundles a set of editing tools: image-to-image transformation, face swap, background removal, image enhancement, and image-to-video conversion, all accessible from a single browser-based interface with no local installation required.
Key Features
- Multiple Flux model tiers — Flux 2 Dev for speed, Flux 2 Pro for quality, Flux 2 Flex for granular control, plus additional models like Seedream and Qwen for complex text rendering.
- LoRA styles library — A curated collection of artistic styles spanning anime, photorealism, and more, selectable at generation time to steer the visual aesthetic without prompt engineering alone.
- Kontext consistency mode — Keeps character appearance and object details coherent across a sequence of generated images, useful for multi-panel or narrative work.
- Bundled editing tools — Face swap, hairstyle modifier, text remover, background changer, professional photo generator, and product photography generator are included alongside the core image generator.
- Image-to-video conversion — Powered by Seedance 1.0 Lite, the platform converts still images into short video clips with configurable duration and resolution settings.
- API access — A RESTful API with webhook support lets developers integrate image generation into their own applications using Python, JavaScript, or any HTTP client.
- Free tier with daily credits — Basic generation is available without a subscription; a credit-based system scales up to commercial-grade output, higher resolutions, and batch processing.
Use Cases
- Content creation — Social media managers and bloggers generate on-brand illustrations or photography-style images from text prompts, bypassing stock photo licensing.
- E-commerce product visuals — Store owners use the product photography generator and background tools to produce clean, professional product shots without a physical studio.
- Concept iteration — Designers test visual directions quickly by generating multiple style variants in parallel before committing to a final look.
- Short-form video assets — Marketers convert finished images into brief animated clips suitable for social ads, reels, or presentation slides.
- Developer integration — Independent developers embed the API into creative apps, chatbots, or internal tools that need on-demand image generation without managing their own model infrastructure.
Who Is It For
Flux-2 Online targets a broad range of creators and builders who need AI-generated visuals without the overhead of running models locally. Graphic designers and marketing professionals will find the multi-model selector and LoRA library useful for iterating on visual styles quickly. E-commerce operators benefit from the dedicated product photography and background editing tools. Developers who want image generation as a service rather than a self-hosted pipeline can connect via the RESTful API. The free daily credit allowance also makes the platform accessible to individual hobbyists and students who want to explore AI image generation before committing to a paid plan.
Tools
list_styles
Return the canonical list of image-generation styles or presets the site exposes. (Flux-2 Online)
Input: no parameters. Returns: text/markdown.
get_pricing
Return the canonical pricing entry point for Flux-2 Online.
Input: no parameters. Returns: text/markdown.
get_official_links
Return the canonical list of official links for Flux-2 Online (website, support, docs when available).
Input: no parameters. Returns: text/markdown.
Resources
site://flux-2-online/styles— Supported image-generation styles and presets.site://flux-2-online/pricing— Canonical pricing entry point.site://flux-2-online/faq— Short FAQ generated from public site metadata.site://flux-2-online/links— Canonical URLs to share with users.
Prompts
tell_me_about_flux_2_online
Summarize what the site is, who it's for, and how it works. — Flux-2 Online
try_image_style_flux_2_online
Recommend a starting image-generation style for a stated goal. — Flux-2 Online
Installation
Install via Smithery
(Replace claude with cursor, windsurf, or continue for those clients.)
Install from source
Then add to your MCP client config (claude_desktop_config.json for Claude Desktop, mcp.json for Cursor / Windsurf / Continue):
Debug with MCP Inspector
Official Links
- Website: https://flux-2.online
- Pricing: https://flux-2.online/pricing
- Support: [email protected]
Development
License
MIT
来源:README.md,提交 c2056de
工具
0版本历史
1- v0.1.0最新Sep 16, 2026
评分
已满足适用检查项的 50%
该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。
- 近 30 天内可连通25 分
请保持端点能够响应 MCP 握手;SourceWeft 每 30 天探测一次。
- 已声明认证方式10 分
请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。
- 已列出工具15 分
请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。
- 描述必需15 分
- 图标10 分
- 文档15 分
- 许可证10 分