# TENET

> Context that compounds. Portable memory, training, and policy for AI agents.

TENET is a CLI that gives AI agents persistent memory and a learning loop. Install it alongside Claude Code, Cursor, or any agent. It remembers what happened, runs improvement experiments overnight, and only keeps changes that pass your evals.

## Install

```
npm install -g @10et/cli
tenet init
tenet setup
```

## Key Commands

```
tenet init          # Add TENET to your project
tenet setup         # Auto-detect stack, create eval agents
tenet status        # Show project health + agent scores
tenet morning       # What happened overnight
tenet build         # Generate build agents from specs
tenet peter         # Run agent orchestrator (RL loop)
tenet ask           # Query project memory
tenet organize      # Check docs against journal evidence
tenet synopsis      # Summarize recent work
```

## Core Concepts

- **Memory**: Persistent knowledge graph — decisions, features, discoveries indexed with embeddings + BM25 search
- **Training Buffer**: (state, action, reward) tuples from every agent action
- **Policy Head**: RL-trained model that predicts which actions improve metrics
- **Eval Loop**: Continuous measurement — agent proposes change, eval measures delta, keep or revert
- **Build Evals**: Spec → decomposed eval → agent iterates 0%→100% → PR created
- **Session Presence**: Real-time awareness of who's working where via Subway P2P mesh
- **Context Hub**: MCP server providing semantic search across project knowledge

## Demo

See the interactive demo showing context compounding across 5 sessions:
https://10et.ai/demo

## Docs

- [Getting Started](https://docs.10et.ai/introduction)
- [Quickstart](https://docs.10et.ai/quickstart)
- [Capabilities](https://docs.10et.ai/capabilities)
- [How It Works](https://docs.10et.ai/how-it-works)
- [Architecture](https://docs.10et.ai/architecture)
- [Agent Harness](https://docs.10et.ai/agents/overview)
- [Build Evals](https://docs.10et.ai/learning/build-evals)
- [CLI: tenet build](https://docs.10et.ai/cli/build)
- [Memory System](https://docs.10et.ai/memory/overview)
- [Context Hub](https://docs.10et.ai/hub/overview)
- [Subway Mesh](https://docs.10et.ai/integrations/subway-mesh)
- [CLI Reference](https://docs.10et.ai/cli/init)

## Links

- Website: https://10et.ai
- Demo: https://10et.ai/demo
- Docs: https://docs.10et.ai
- GitHub: https://github.com/10et-ai/cli
- npm: https://www.npmjs.com/package/@10et/cli