AITree MCP Server
Enables AI assistants to generate annotated file trees, retrieve file statistics, list git-changed files, and read file contents from local directories or GitHub repositories.
README
AITree
Project file map generator for AI-assisted development.
Scans a directory and prints a clean, annotated tree that you can paste directly into any AI chat. Instead of uploading your entire codebase, share the map first — then only send the files the AI actually asks for.
Live Demo: paracci.github.io/aitree
webapp/
├── src/
│ ├── components/
│ │ ├── Header.jsx [4KB]
│ │ └── Footer.jsx [2KB]
│ ├── App.jsx [8KB]
│ └── index.js [1KB]
├── public/
│ └── index.html [3KB]
└── package.json [1KB]
# 2 directories, 6 files (18KB total)
#
# File types:
# .jsx 3 ███
# .js 1 █
# .html 1 █
# .json 1 █
Installation
Via pip (recommended):
pip install paracci-aitree
One-liner (Linux / macOS):
curl -fsSL https://raw.githubusercontent.com/paracci/aitree/main/install.sh | bash
One-liner (Windows PowerShell):
irm https://raw.githubusercontent.com/paracci/aitree/main/install.ps1 | iex
From source:
git clone https://github.com/paracci/aitree.git
cd aitree
pip install .
Optional dependencies — install only what you need:
pip install "paracci-aitree[live]" # --live mode (watchdog)
pip install "paracci-aitree[tokens]" # --tokens (tiktoken)
pip install "paracci-aitree[git]" # --git-changed (gitpython)
pip install "paracci-aitree[mcp]" # MCP server (mcp)
pip install "paracci-aitree[all]" # everything + clipboard (pyperclip)
Usage
aitree [path] [options]
path defaults to the current directory if omitted.
Output modes
| Flag | Description |
|---|---|
| (none) | Print tree to stdout |
--save |
Save to _aitree.txt (or _aitree.json / _aitree.md) |
--copy |
Copy output to clipboard |
--live |
Watch for changes and auto-update the saved file |
--serve |
Launch the web UI in your browser |
Output formats
| Flag | Description |
|---|---|
--format text |
Plain text with # comments (default) |
--format json |
Structured JSON with tree + stats + meta |
--format markdown |
Markdown with fenced code block + stats table |
Filtering
| Flag | Description |
|---|---|
--depth N |
Limit tree depth (e.g. --depth 2) |
--include GLOB |
Show only matching files (repeatable) |
--exclude GLOB |
Hide matching files (repeatable) |
--no-gitignore |
Skip .gitignore / .aitreeignore parsing |
--git-changed |
Show only git-changed / untracked files |
Extras
| Flag | Description |
|---|---|
--tokens |
Append estimated token count to output |
--version |
Show version and exit |
Examples
aitree . # current directory → stdout
aitree ~/projects/webapp # specific path → stdout
aitree . --save # save to _aitree.txt
aitree . --copy # copy to clipboard
aitree . --depth 2 # top 2 levels only
aitree . --include "*.py" # Python files only
aitree . --include "*.py" --include "*.js" # multiple patterns
aitree . --exclude "*.test.*" # hide test files
aitree . --git-changed # only modified/untracked files
aitree . --format json --save # save as _aitree.json
aitree . --format markdown --save # save as _aitree.md
aitree . --tokens # append token count
aitree . --live # watch for changes
aitree . --live --depth 2 --git-changed # live + depth + git filter
aitree . --serve # open web UI
aitree . --serve --port 8080 # web UI on a custom port
AI workflow
1. aitree . --copy
2. Paste into your AI chat
3. Ask "Which files do you need?"
4. Send only those files
Web UI
--serve launches a local web server and opens a professional, interactive explorer in your browser.
- Real-time Search: Instant tree filtering with match highlighting.
- Syntax Highlighting: Auto-detection for 30+ languages (via highlight.js).
- Themes: Switch between Dark (Tokyo Night) and Light (GitHub) modes.
- File Preview: Click any file to see content with line numbers and word-wrap.
- Interactive Tree: Fast collapse/expand, depth control, and file-type icons.
- Live Reload: UI auto-refreshes when your project files change.
- Responsive Layout: Resizable sidebar and mobile-friendly design.
aitree . --serve
aitree . --serve --port 8080
MCP Server
AITree ships a Model Context Protocol server so AI assistants (Claude Desktop, etc.) can call it as a tool.
# Install MCP dependency
pip install "paracci-aitree[mcp]"
# Start the server
aitree-mcp
Available tools
| Tool | Description |
|---|---|
aitree_get_tree |
Generate a file tree for a local path or GitHub URL |
aitree_get_stats |
Return structured JSON statistics for a path |
aitree_get_changed |
List git-changed / untracked files |
aitree_read_file |
Read the contents of a file inside the project |
aitree_read_file lets an AI assistant fetch any file it spotted in the tree — without leaving the conversation:
aitree_get_tree(path=".", format="text")
aitree_read_file(path=".", file="src/auth/login.py")
Claude Desktop config (claude_desktop_config.json)
{
"mcpServers": {
"aitree": {
"command": "aitree-mcp"
}
}
}
GitHub URL support
Pass a GitHub repository URL instead of a local path:
aitree https://github.com/owner/repo
aitree https://github.com/owner/repo --depth 2
aitree https://github.com/owner/private-repo --token ghp_xxxx
Works entirely via the GitHub API — no cloning required.
Custom ignore rules
Create .aitreeignore in your project root — same syntax as .gitignore:
# .aitreeignore
*.secret
config/local.json
private/
AITree reads .aitreeignore first, then .gitignore. Use --no-gitignore to skip both.
What gets ignored by default
| Category | Examples |
|---|---|
| VCS & editors | .git, .idea, .vscode, .svn |
| Dependencies | node_modules, .venv, venv, env |
| Build output | dist, build, target, out, bin |
| Cache | __pycache__, .pytest_cache, .mypy_cache |
| Binary & temp | .pyc, .class, .exe, .log, .tmp |
| OS files | .DS_Store, Thumbs.db |
Live mode
--live watches the directory with watchdog and regenerates the output file whenever the structure actually changes. False positives (editor auto-saves, antivirus touches, etc.) are filtered out via content hashing and a 1-second debounce.
🟢 AITree LIVE — Watching: /home/user/projects/webapp [depth=2, git-changed]
Updates trigger only on real changes.
Press Ctrl+C to stop.
[14:32:10] Updated — initial snapshot
[14:33:05] Updated — filesystem change (3 false events skipped)
Combines with any filter flag: --depth, --include, --exclude, --git-changed.
Platform support
| Platform | Tested |
|---|---|
| Linux | ✅ |
| macOS | ✅ |
| Windows | ✅ |
| WSL | ✅ |
Requires Python 3.10+.
License
Recommended Servers
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.
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.
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.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
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.
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.