I built Web MCP Bridge a 100% free and open-source extension (MIT) that connects web-based AI interfaces (ChatGPT, Claude, Gemini, DeepSeek & more) directly to your local Model Context Protocol (MCP) servers.
No API subscriptions or paid plans needed just bring your local tools, files, and databases straight into the web UIs you already use every day.
Already available on GitHub! Feel free to check out the repo or start using it right away: https://github.com/AstraxOf/web-...
Model Context Protocol (MCP) servers allow AI to access local files, databases, and custom tools. Web MCP Bridge brings this capability directly to web interfaces like ChatGPT, Claude, Gemini, DeepSeek and more—completely free and 100% open source under MIT. No API subscriptions needed.
AI agents are getting pretty good at finding information, summarizing documents, and answering questions. But what about letting them change the information itself?
Imagine an AI agent finds outdated onboarding instructions in your company's wiki. Should it be allowed to fix them immediately? Suggest an edit for someone to review? Or simply flag the problem and leave everything to a human?
We've been thinking about this while developing the XWiki MCP Server.
Our approach is to respect existing user permissions and keep write operations disabled by default. Administrators can enable them when needed.
