Welcome to OWL
OWL is a local AI daemon with persistent memory, project-scoped context, and an evolving soul. Unlike stateless AI assistants, OWL remembers your preferences, learns from interactions, and develops a unique personality over time.
(o,o)
/)_) OWL - A wise AI with a growing soul
" "
What Makes OWL Different?
Persistent Memory
OWL maintains three layers of memory:
- Memory File: Human-readable facts and preferences (
~/.owl/memory.md) - SQLite Database: Conversation history and learnings
- Session Summaries: Compressed context for long conversations
Project Awareness
OWL automatically detects your project type and framework, providing context-aware assistance without repeated explanations.
Evolving Soul
Unlike static AI personas, OWL's character evolves based on interactions. It reflects on conversations, extracts learnings, and develops a unique personality.
Local & Private
Everything runs locally on your machine:
- Ollama for LLM inference
- ChromaDB for knowledge retrieval
- SQLite for persistent storage
No data leaves your machine.
Quick Example
# Start OWL (auto-starts daemon if not running)
owl
# Chat naturally
you: What files are in this project?
owl: [Uses list_files tool and provides response]
you: Remember that I prefer tabs over spaces
owl: I'll remember your preference for tabs over spaces.
# Your preference is now stored permanently
You can also manage the daemon directly:
owl status # Check if daemon is running
owl stop # Stop the daemon
owl start # Start daemon in background
Architecture Overview
┌─────────────────────────────────────────────────────────────┐
│ OWL Daemon │
│ ┌─────────┐ ┌─────────┐ ┌─────────┐ ┌─────────────────┐│
│ │ LLM │ │ Memory │ │Knowledge│ │ Soul ││
│ │ (Ollama)│ │(SQLite) │ │(ChromaDB│ │(YAML + Evolver) ││
│ └────┬────┘ └────┬────┘ └────┬────┘ └────────┬────────┘│
│ └───────────┴───────────┴─────────────────┘ │
│ │ │
│ Unix Socket │
│ │ │
└───────────────────────────┼─────────────────────────────────┘
│
┌───────┴───────┐
│ OWL CLI │
│ (Interactive │
│ REPL) │
└───────────────┘
Next Steps
- Installation Guide - Get OWL running on your machine
- Quick Start Tutorial - Learn the basics in 5 minutes
- CLI Commands - Reference for all available commands
- Architecture Overview - Deep dive into how OWL works