
Deploy Engram
Self-hosted MCP second brain: long-term memory for your AI agents
engram
Just deployed
/data
Deploy and Host Engram on Railway
Engram is a self-hosted MCP server and dashboard that gives Claude Code, Cursor, Hermes, and any Model Context Protocol agent shared, long-term memory — over a plain, git-backed folder of Markdown. Humans edit notes in a fast dashboard; agents read and write the same knowledge through one MCP endpoint.
About Hosting Engram
Engram runs as a single container with a persistent volume mounted at /data, where it keeps app state and clones of your Markdown vault. It is not serverless — it holds a file watcher, an in-memory search + knowledge-graph index, and an optional git-sync loop, so it needs a long-running process and a disk. This template provisions the service and volume for you; you supply a handful of bootstrap variables (an auth secret, your public URL, an email allowlist, and Google OAuth credentials for dashboard login). Everything else — git sync, AI capture, GitHub connect, app name — is configured afterward in the in-app Settings page, with no redeploy.
Common Use Cases
- Long-term memory for Claude Code, Cursor, and Hermes agents — stop re-explaining your project every session.
- A shared team knowledge base that AI agents can read and write over MCP.
- A self-hosted, Obsidian-compatible second brain — your notes, your server, your git repo.
Dependencies for Engram Hosting
- A GitHub repository of Markdown notes to serve as the vault (connected in the dashboard after deploy).
- A Google OAuth app for dashboard login.
Deployment Dependencies
- Google Cloud Console (OAuth credentials): https://console.cloud.google.com/apis/credentials
- Model Context Protocol: https://modelcontextprotocol.io
- Source & full setup guide (DEPLOY.md): https://github.com/klyiro/engram
Implementation Details
After deploy, mount a volume at /data, then connect your vault repo in Workspaces and create MCP tokens on the Connect page. Point agents at https:///api/mcp — Claude Code, Cursor, and Cline authenticate with a bearer token; Claude.ai web connects via the built-in OAuth flow (enabled automatically once AUTH_SECRET is set). Required variables: AUTH_SECRET, APP_URL, ALLOWED_EMAILS, GOOGLE_CLIENT_ID, GOOGLE_CLIENT_SECRET, and ENGRAM_DATA_DIR=/data.
Why Deploy Engram on Railway?
Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.
By deploying Engram on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.
Template Content
engram
rwnalds/engramALLOWED_EMAILS
Comma-separated (test@gmail.com,another@gmail.com))
GOOGLE_CLIENT_ID
GOOGLE_CLIENT_SECRET
