zanfiel

zanfiel / engram

Public

Persistent memory for AI agents

engram.lol agents ai bun embeddings knowledge-graph
19
2
100% credibility
Found Mar 10, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Engram is a self-hosted memory system for AI agents that stores, organizes, searches, and recalls user facts, decisions, and context with smart linking and categories.

How It Works

1
📖 Discover Engram

You hear about Engram, a helpful notebook that lets your AI friends remember important details forever, like facts, decisions, and preferences.

2
🖥️ Start it up

You download and launch Engram on your computer with a simple one-click setup, and it runs quietly in the background.

3
🔐 Set your private password

You pick a secure password to keep all your memories safe and private, just for you.

4
💾 Save your first memory

You tell Engram to remember something special, like 'I prefer dark mode' or a key decision, and mark how important it is.

5
🔍 Find and recall memories

You ask Engram about something from before, and it instantly pulls up the perfect matching memories with details on why they matter.

6
🤖 Connect to your AI assistant

You link Engram to your favorite AI chat tool, so it feeds in remembered context automatically for smarter conversations.

🎉 AI remembers everything

Now your AI companion never forgets you or past talks, making every interaction feel personal and helpful.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 19 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is engram?

Engram is a self-hosted TypeScript server providing persistent memory for AI agents, storing facts, decisions, and context with automatic semantic embedding, linking, and decay. You get CLI commands like `engram store`, `search`, `recall`, and `sync`, plus Python/TS SDKs and Docker setup for local github persistent storage on port 4200. It solves stateless LLM sessions by enabling smart retrieval—static facts, recent activity, high-importance items—packaged for token budgets.

Why is it gaining traction?

MCP server integrates directly with Claude Desktop, Cursor, or Windsurf for zero-config tool calls like engram_store or engram_recall, outpacing generic RAG libs. Bi-directional sync and project/entity scoping beat basic key-value stores, while OpenAPI spec simplifies engram deepseek or custom agent flows. Local-first design hooks devs ditching cloud for engram gmbh-style control.

Who should use this?

AI agent builders needing cross-session recall for chatbots or tools, like tracking user prefs in multi-turn convos. Suited for TypeScript/Node devs prototyping engram bremen prototypes or Python scripters wanting engrams without engramme ark complexity. Skip if you're deep into haskell persistent github or iptables-persistent github stacks.

Verdict

Solid for local persistent memory experiments—API, CLI, and MCP make it instantly usable despite 11 stars and 1.0% credibility score. Early maturity means watch for stability, but deploy now if self-hosted agent context is your blocker.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.