MuninnDB

Installation

Install Script

The quickest way to install MuninnDB on macOS or Linux:

bash 
curl -fsSL https://muninndb.com/install.sh | sh

The install script detects your platform (macOS arm64/amd64, Linux amd64/arm64), downloads the correct binary, and places it in /usr/local/bin/muninn.

Windows (PowerShell)

On Windows, run PowerShell and use the official install script:

PowerShell 
irm https://muninndb.com/install.ps1 | iex

This downloads the latest release, extracts to %LOCALAPPDATA%\muninn, and adds it to your PATH. If muninn start shows an ORT/DLL error (bundled embedder), install the Visual C++ Redistributable (x64) — see Troubleshooting.

Homebrew

macOS users can install via Homebrew. The one-liner taps and installs in one step:

bash 
brew install scrypster/tap/muninn

# To update later:
brew upgrade muninn

Configuration

MuninnDB works with zero configuration out of the box. Settings are configured via CLI flags or environment variables — no config file required.

bash — key flags 
muninn start --data /path/to/data    # Custom data directory (default: ~/.muninn/data)
muninn start --mcp-token my-token    # MCP bearer token
muninn start --mcp-addr :9000        # Custom MCP port
muninn start --log-level debug       # Log verbosity: debug, info, warn, error

# Embedder (optional — enables semantic similarity):
MUNINN_OLLAMA_URL=ollama://localhost:11434/nomic-embed-text muninn start
MUNINN_OPENAI_KEY=sk-...  muninn start

# LLM enrichment (optional — auto-summaries and entity extraction):
MUNINN_ENRICH_URL=anthropic://claude-haiku-4-5-20251001 \
MUNINN_ANTHROPIC_KEY=sk-ant-... muninn start

First Run

For first-time setup, run muninn init — it walks through connecting your AI tools and starts all services automatically:

bash 
muninn init     # first-time guided setup
muninn start    # start all services in background
muninn status   # check which services are running
muninn stop     # stop the running server
muninn shell    # interactive memory explorer

# Ports:
# MBP  :8474   binary protocol
# REST :8475   JSON API
# gRPC :8477   gRPC API
# MCP  :8750   AI tool integration (http://localhost:8750/mcp)
# UI   :8476   web dashboard (http://localhost:8476)
← Previous Next →