What is Engrammic?
Persistent memory for AI agents
Engrammic gives your AI coding agents memory that does not rot, does not hallucinate, and can show its work across sessions.
How It Works
Your agent stores observations, learnings, and beliefs in a knowledge graph. Each piece of knowledge tracks its provenance: where it came from and why the agent believes it. When you start a new session, your agent can recall relevant context from past work.
Cross-session memory: preferences, decisions, and codebase knowledge persist between conversations.
Provenance tracking: ask "why do you believe that?" and get an evidence chain back to source.
Shared memory: multiple agents can build on each other's knowledge.
Get Started
Ready to add memory to your agent? Head to join.engrammic.ai, one click for VS Code and Cursor, copy-paste config for other tools, or one command that sets up everything.
After Install
Once connected, verify it's working:
"What MCP tools do you have?"
Your agent should list Engrammic tools like remember, learn, recall, and trace.
Then try storing something:
"Remember that I prefer TypeScript over JavaScript"
Close your session, reopen, and ask:
"What do you know about my preferences?"
Your agent recalls it from the previous session.