kael-bit

kael-bit / engram-rs

Public

Memory engine for AI agents — time axis (3-layer decay/promotion) + space axis (self-organizing topic tree). Hybrid search, LLM consolidation. Single Rust binary.

agents ai ai-agents ai-memory claudecode
16
1
100% credibility
Found Feb 27, 2026 at 12 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

engram-rs provides persistent, brain-inspired memory for AI agents with automatic organization into time-based layers and topic clusters.

How It Works

1
🔍 Discover engram

You hear about engram, a simple way to give your AI a real memory that remembers like a brain.

2
📥 Get it running

Run a quick download script that guides you through setup on your computer.

3
🔗 Link your AI helper

Tell it which smart service to use, like your favorite AI chat, so it can think deeply.

4
🚀 Start your memory bank

Click to launch, and your personal memory space comes alive on your local machine.

5
🧠 Feed it conversations

Share chats or notes with your AI, and watch it automatically sort important bits into short-term, active, and forever memories.

6
🔍 Ask and remember

Query your memory anytime, and get smart, organized results that connect related ideas.

✨ Your AI grows wiser

Over time, your AI recalls lessons, preferences, and decisions perfectly, feeling more like a trusted partner.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 12 to 16 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-rs?

engram-rs is a bright memory engine for AI agents, packaged as a single Rust binary under 10MB. It stores facts, decisions, and lessons in a brain-like system: buffer for fresh intake with automatic decay, working for active context, and core for permanent identity, all with promotion based on access and quality. Retrieve via hybrid semantic/keyword search or browse a self-organizing topic tree, using simple HTTP endpoints like POST /memories and GET /recall.

Why is it gaining traction?

Stands out from basic vector DBs by enforcing memory lifecycle—unused entries fade, duplicates merge via LLM, and topics cluster automatically—cutting noise in agent loops. The curl installer, Docker support, and MCP tools for Claude/Cursor make it a seamless github memory manager add-on, with background consolidation skipping idle periods for efficiency.

Who should use this?

AI agent builders handling session continuity, like memory engineers integrating recall into LLM chains. Ideal for Rust devs prototyping autonomous tools, or Cursor/Claude users bypassing github memory limits with /resume for context bootstrap and /triggers for safety checks before deploys.

Verdict

Promising memory engine ai at 1.0% credibility and 11 stars—great docs and installer, but light on tests signals early maturity. Try the binary now if agent memory is your bottleneck; stabilize for production.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.