nagisanzenin/engram
1,151 stars · Last commit 2026-07-29
Evidence-based learning engine for Claude Code — first-principles curricula, free-recall verification with receipts, FSRS-scheduled memory, and explorable artifacts. Learn anything; keep it.
README preview
<p align="center"> <img src="assets/banner.png" alt="Engram — learn anything. keep it." width="100%"> </p> <p align="center"> <img src="https://img.shields.io/badge/version-1.10.1-6D4AA8.svg" alt="Version 1.10.1"> <a href="https://www.npmjs.com/package/opencode-engram-learning"><img src="https://img.shields.io/npm/v/opencode-engram-learning?label=npm&color=6D4AA8" alt="npm package"></a> <img src="https://img.shields.io/badge/license-MIT-yellow.svg" alt="MIT License"> <img src="https://img.shields.io/badge/selftest-302%2F302-3E7D5A.svg" alt="302/302 checks"> <a href="gold/assessor-gold.jsonl"><img src="https://img.shields.io/badge/grader%20inflations-0%2F258-3E7D5A.svg" alt="0 of 258 blind judgments graded up, on the shipping spec"></a> <img src="https://img.shields.io/badge/scheduler-FSRS--4.5-6D4AA8.svg" alt="FSRS-4.5"> <a href="CONTRIBUTING-DATA.md"><img src="https://img.shields.io/badge/data-100%25%20local-3E7D5A.svg" alt="100% local — the engine has no network code, proven by a permanent selftest"></a> <a href="https://discord.gg/temm1e"><img src="https://img.shields.io/badge/discord-community-5865F2.svg" alt="Discord community"></a> </p> <h3 align="center">Your AI can explain anything. Engram makes sure <em>you</em> still know it next month.</h3> > **The mix-up worth clearing first: Engram is not an agent-memory plugin.** It doesn't give your agent persistent memory, context, or knowledge of your codebase — memory MCPs and context tools do that, *for the agent*. Engram points the other way: **it's a learning system for the human.** Your agent becomes a tutor that makes you do the thinking, a blind examiner that checks you actually got it, and a scheduler that brings each idea back right before your brain drops it. The agent doesn't get smarter. **You do — measurably, with receipts.** Born as a Claude Code plugin; the same skills and engine now run on six agentic platforms — including, as of v1.0.8, one that puts the tutor in your chat app: