Getting Started with Infigraph
Quick Start
Install (one-liner)
macOS / Linux:
curl -fsSL https://raw.githubusercontent.com/intuit/infigraph/main/install.sh | bash
Windows (PowerShell):
iwr https://raw.githubusercontent.com/intuit/infigraph/main/install.ps1 -UseBasicParsing | iex
Index your project
cd /path/to/project
infigraph index
Ask your AI agent
"Who calls the validate_user function?"
"Show me the blast radius of this change"
"Find authentication logic in this codebase"
"What's the architecture of this project?"
Infigraph auto-indexes on first query. No manual setup needed for Claude Code, Cursor, or other AI agents with MCP support.
How It Works
With Claude Code (Recommended)
After install, start using Claude Code normally. Infigraph indexes your project automatically and transparently:
> You: "Search for authentication logic"
> Claude: [Scans via Infigraph, returns precise results with 60-80% fewer tokens]
No CLI commands. No separate indexing step. Just ask.
With Other AI Agents
Any agent with MCP support (Cursor, VS Code + Copilot, Windsurf, etc.) gets 69 Infigraph tools automatically after infigraph install.
Manual CLI (Optional)
infigraph search "auth" # Hybrid search
infigraph query "MATCH (...)" # Cypher queries
infigraph trace-callers "function_name" # Who calls this?
infigraph dead-code # Find unused functions
infigraph impact "auth.py::authenticate" # Blast radius
Offline-First Design
Infigraph is built for offline operation — everything runs locally, no cloud APIs or network access needed. The ML embedding model (potion-base-8M, 29MB) is bundled in this repository for immediate use without additional downloads.
This means:
- Semantic search works out of the box after cloning
- No external dependencies or API keys required
- Your codebase never leaves your machine
- Works on air-gapped systems
Installation
System Requirements
| Platform | Required |
|---|---|
| macOS | Rust (rustup), brew install cmake |
| Linux | Rust (rustup), sudo apt install cmake |
| Windows | Rust (rustup), Visual Studio Build Tools (C++20) |
No Docker, no Python, no Node.js required — everything is self-contained.
macOS / Linux
curl -fsSL https://raw.githubusercontent.com/intuit/infigraph/main/install.sh | bash
This:
- Downloads pre-built binaries from GitHub releases (if available)
- Falls back to cloning +
cargo build --release(installs Rust if needed) - Adds
infigraph,infigraph-mcp, andlsp-to-scipto~/.local/bin - Registers MCP server for all 11 AI coding agents
- Writes primary search instructions to
~/.claude/CLAUDE.md
System dependency:
cmakeis required to build the graph database. Install before building:brew install cmake(macOS) orsudo apt install cmake(Linux).
Windows
Run this single command from PowerShell:
iwr https://raw.githubusercontent.com/intuit/infigraph/main/install.ps1 -UseBasicParsing | iex
This downloads and runs the full installer — which fetches the pre-built binary and registers the MCP server.
Update
Re-run the installer to pull latest and rebuild:
curl -fsSL https://raw.githubusercontent.com/intuit/infigraph/main/install.sh | bash
Or if building manually:
cd /path/to/infigraph && git pull && cargo build --release
infigraph update
infigraph update re-registers MCP server paths and refreshes CLAUDE.md instructions.
Infigraph also checks for updates in the background (once per 24h) and prints a hint when a newer version is available.
Uninstall
infigraph uninstall
Removes:
- MCP server config from all 11 AI agents
- Primary search instructions from
~/.claude/CLAUDE.md
Does NOT delete the binary — remove ~/.local/bin/infigraph and ~/.local/bin/infigraph-mcp manually if desired.