eclipse-chatgpt-plugin

MCP.Pizza Chef: gradusnikov

An Eclipse plugin with two halves. One is a chat panel in the editor that talks to OpenAI, Anthropic, Groq, DeepSeek, or a model running on your own machine, and can be pointed at other connectors you add yourself. The other opens Eclipse up to outside agents such as Claude Code, which can then edit files, run refactorings, launch builds, run tests, step through the debugger, and commit to Git. You bring your own model account.

Coding

Use This MCP client To

Ask about the class I have open without pasting any code Rename a Java type everywhere it is used Run the failing test and tell me what broke Fix the compile errors showing in my workspace Set a breakpoint and walk me through the values Stage and commit my changes with a decent message Restore a file to the version from before lunch

README

AssistAI - Eclipse IDE as an MCP Server for AI Agents

AssistAI is an Eclipse IDE plugin that exposes your entire development environment as an MCP (Model Context Protocol) server. External AI agents -- Claude Code, OpenAI Codex, Claude Desktop, or any MCP-compatible client -- can read, navigate, edit, build, test, run, and debug your Java projects directly through Eclipse, preserving workspace sync, local history, and incremental compilation.

AssistAI also includes a built-in LLM chat view for quick inline interactions with any supported model.

Why MCP through Eclipse?

When AI agents edit files through the filesystem directly, Eclipse doesn't know anything changed. Editors show stale content, incremental compilation misses updates, and local history gaps appear.

AssistAI solves this by routing all operations through Eclipse APIs:

  • Edits go through JDT -- incremental compilation fires immediately, errors update in real time
  • Refactorings use Eclipse's refactoring engine -- renames, moves, and package restructures update all references across the workspace
  • File reads reflect the editor buffer -- agents always see the latest unsaved content, not the on-disk version
  • Local history is preserved -- every change is tracked, undoable through Eclipse's local history
  • Tests run inside Eclipse -- JUnit results, console output, and compilation errors are accessible as tool responses

Getting Started with External Agents

1. Enable the HTTP MCP Server

  1. Open Window > Preferences > Assist AI > HTTP MCP Server
  2. Check Enable HTTP MCP Server
  3. Set Hostname and Port (defaults: localhost:8124)
  4. Click Generate to create an authentication token
  5. Click Apply -- the server starts immediately

The status panel shows all available endpoints:

  • http://localhost:8124/mcp/eclipse-ide -- code analysis, navigation, search, testing, builds
  • http://localhost:8124/mcp/eclipse-coder -- file editing, refactoring, patching, formatting
  • http://localhost:8124/mcp/eclipse-runner -- launch, debug, breakpoints, stepping
  • http://localhost:8124/mcp/eclipse-context -- resource cache, file local history, version tracking
  • http://localhost:8124/mcp/eclipse-git -- git status, log, staging, commit, diff, branches, stash

2. Connect Your Agent

eclipse-chatgpt-plugin FAQ

Do I need a paid account?
For the built-in chat panel, yes — your own OpenAI, Anthropic, Groq, or DeepSeek account, or a model running locally on your machine, which costs nothing. For the other direction, where an outside agent drives Eclipse, no model account is needed here because the agent brings its own.
Can I use this to let Claude Code work inside my Eclipse projects?
Yes. Switch on the built-in server in preferences, click generate to make a token, then paste a short config block into your agent. It also works with Codex and Claude Desktop.
What can it actually change?
A great deal. It can create, overwrite, rename, and delete files, run search-and-replace across the whole workspace, launch and debug your programs, run Maven builds, and stage, commit, branch, or delete branches in Git. There is no read-only setting. Eclipse's own local history is your safety net, and it can restore old versions on request.
Is it safe to leave switched on?
The built-in server stays on your own machine and demands a token you generate. Anything holding that token has all the reach described above, so turn the server off when you are not using it.
Does it only work with Java?
The deep parts do. Refactoring, type hierarchies, test running, and debugging are Java-specific. File editing, search, Git, console output, and document navigation work on anything in the workspace.
Why go through Eclipse instead of letting the agent edit files directly?
Because Eclipse then knows the change happened. Errors update as you watch, renames update every reference, unsaved editor content is what the agent sees, and every edit lands in local history.
How hard is setup?
The plugin itself installs from the Eclipse Marketplace with a drag. Connecting an outside agent is a checkbox, a generated token, and a pasted config block.
Does it cost anything?
The plugin is free and open source. You pay whatever your chosen model provider charges, or nothing at all if you point it at a model running locally.