Prerequisites

Requires Node.js 20+. We recommend installing via nvm:

curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.3/install.sh | bash
nvm install 22

Install

npm install -g @10et/cli

Quick Start

# Add to your project
cd your-project
tenet init

# Set up improvement agents
tenet setup

# Run overnight improvement loop
tenet peter autoresearch --rounds 5

# Check results in the morning
tenet morning

What It Does

TENET gives AI agents persistent memory and a learning loop. Install it alongside Claude Code, Cursor, or any agent. Your codebase improves while you sleep.

Memory — Decisions, patterns, and outcomes persist across sessions. Agents don't start from zero.

Learning Loop — Every action produces (state, action, reward) tuples. A policy head learns what works for YOUR codebase.

Overnight Agents — RL agents try improvements while you sleep. Eval measures the delta, keep what improves, revert what doesn't.

Build Agents — Write a spec, tenet build generates the eval. Agents iterate from 0% → 100%. PR created automatically.

Skills — Reusable instructions that get better through usage. Browse the skill registry, or create your own.

Context Hub — MCP server any agent connects to. Semantic search across your entire project history.

Commands

| Command | What it does | |---------|-------------| | tenet init | Add TENET to your project | | tenet setup | Auto-detect stack, create eval agents | | tenet morning | What happened overnight | | tenet build | Generate build agents from specs | | tenet peter | Run agent orchestrator | | tenet ask | Query project memory | | tenet skills list | Browse installed skills | | tenet status | Project health dashboard |

How It Compounds

Day 1:     tenet init — first session journaled
Week 1:    Suggestions reference previous work
Month 1:   Agents run overnight — PRs waiting at breakfast
Month 3:   Ask about any past decision, get the full context

Works With

Any agent runtime — Claude Code, Cursor, Pi, OpenClaw. TENET is the context layer underneath.

Links

• Website: 10et.ai
• Demo: 10et.ai/demo
• Docs: docs.10et.ai
• Skills: github.com/10et-ai/skills

License

MIT