ai-brainmap
Live Context Visualizer for Claude Code
See what Claude is thinking — in real time, in 2D or 3D.
ai-brainmap turns Claude Code's invisible reasoning into a living mindmap. Every
user prompt, tool call and response is rendered on the fly, so you can watch
your AI work instead of guessing what it's doing.
Open source. Lightweight. 100 % local. Built for developers who want clarity, not
black boxes.
Install
/plugin marketplace add Austro-Intelligence/ai-brainmap
/plugin install ai-brainmap@austro-intelligence
Then in any Claude Code session:
/brainmap start # start server + open browser
/brainmap stop # stop server
/brainmap status # is it running?
/brainmap clear # reset the map
The server listens on http://127.0.0.1:4823. Set MINDMAP_PORT to change the port.
How it works
┌─────────────────┐ hook payload ┌────────────────────┐
│ Claude Code │ ─────────────────> │ mindmap-hook.py │
│ (your session) │ │ (tiny HTTP client) │
└─────────────────┘ └─────────┬──────────┘
│ POST /api/thought
▼
┌────────────────────┐
│ mindmap-server.py │
│ (localhost:4823) │
│ in-memory state │
└─────────┬──────────┘
│ GET /api/state (polling)
▼
┌────────────────────┐
│ public/index.html │
│ Three.js + D3 │
│ force-graph in the │
│ browser │
└────────────────────┘
- A local Python HTTP server keeps state and serves the Three.js + D3 frontend.
- Claude Code hooks (
UserPromptSubmit, PreToolUse, PostToolUse, Stop)
push events to the server.
- The browser polls
/api/state every 500 ms and updates the force graph
in place.
- Active nodes pulse yellow; finished nodes settle into place.
Repository layout
ai-brainmap/
├── .claude-plugin/
│ └── marketplace.json # Marketplace manifest
├── plugins/
│ └── ai-brainmap/
│ ├── .claude-plugin/plugin.json
│ ├── bin/
│ │ ├── mindmap-server.py # HTTP server + state API
│ │ └── mindmap-hook.py # Hook bridge (stdin → POST)
│ ├── commands/brainmap.md # /brainmap slash command
│ ├── hooks/hooks.json # Claude Code hooks
│ ├── public/index.html # Three.js + D3 frontend
│ └── README.md
├── LICENSE
└── README.md
Requirements
- Claude Code
- Python 3 (preinstalled on macOS/Linux; use the official installer or WSL on Windows)
- A modern browser with WebGL (Chrome, Safari, Firefox, Edge)
Privacy & security
- Fully local. The Python server binds to
127.0.0.1:4823 only. There are
no outbound network calls from the server or the frontend.
- No telemetry. Your prompts and tool calls never leave your machine.
- Hook failure is silent. If the server is not running the hook bridge
returns immediately — Claude Code is never blocked by the plugin.
- Read-only of your own transcript. An optional replay endpoint
(
POST /api/replay) reads a Claude transcript file you pass in and
visualizes its history; it does not access arbitrary files.
Customize
- Port: set
MINDMAP_PORT in the environment.
- Hook filter: edit
plugins/ai-brainmap/hooks/hooks.json — narrow the
matcher to specific tools if you want a quieter map.
- Look & feel: everything lives in
plugins/ai-brainmap/public/index.html
(Three.js scene graph, D3 force parameters, CSS).
Contributing
Issues and PRs welcome. To hack locally, clone the repo and register it as a
local marketplace:
/plugin marketplace add /path/to/ai-brainmap
/plugin install ai-brainmap@austro-intelligence
Then edit files under plugins/ai-brainmap/ and reload the browser — changes
in public/index.html are picked up on refresh.
License
MIT — see LICENSE.
xrf9268-hue
uppinote20
xhyqaq