# nagisanzenin/engram

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.

Repository: https://github.com/nagisanzenin/engram
Canonical: https://ross.abutalabs.com/products/nagisanzenin-engram
Language: Python
License: MIT
License Family: permissive
Topics: claude-code, claude-code-plugin, education, fsrs, learning, learning-science, spaced-repetition
Last push: 2026-08-18T06:13:31+00:00

## Health v2 (maintenance only)
Score: 79/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 98, longevity 4
- inputs: {"age_days": 59, "days_push": 15, "days_rel": 15, "gap_med": 0.0, "n_releases_24m": 53}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1374, forks 101 (observed 2026-08-28T04:04:33.000924+00:00)

## What it is
Engram is a Claude Code plugin (portable to other agentic platforms) that turns your AI coding agent into a personal tutor for the human user. It builds first-principles curricula, verifies learning through free-recall questioning with receipts, and schedules reviews using the FSRS spaced-repetition algorithm — all fully local with no network code.

## Use cases
- learn a new topic with an AI tutor that makes me do the thinking
- verify I actually understood something my agent explained
- schedule spaced-repetition reviews of concepts I learned
- build a first-principles curriculum for any subject
- test my recall with a blind examiner instead of re-reading notes
- keep knowledge long-term instead of forgetting it next month
- use my coding agent as a learning coach

## When to choose
- you use Claude Code, Codex, OpenCode, or another supported agentic platform and want to learn and retain material
- you want evidence-based learning with FSRS scheduling and verifiable recall checks
- you need a fully local, privacy-preserving learning engine with no network calls
- you want measurable proof of what you actually know

## When to avoid
- you want your agent to have persistent memory or codebase context — this is not an agent-memory tool
- you need a standalone flashcard app without an AI agent
- you want a hosted or cloud-synced learning platform

## Facets
- artifact type: plugin
- maturity: active
- function: scheduling, nlp, developer-tools
- domain: education, artificial-intelligence, developer-tools
- platform: cli, cross-platform, python
- tags: claude-code-plugin, spaced-repetition, fsrs, learning-science, free-recall, tutor, agentic-platforms, local-first, education

## Member repositories
- nagisanzenin/engram (main) score 79

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:33.000924+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:40:38.322536+00:00, confidence not recorded.
  - readme: https://github.com/nagisanzenin/engram (fetched 2026-08-28T04:04:33.000924+00:00, sha ace01760c5c7)
- Data as of 2026-08-30T08:39:29.467469+00:00.