Getting Started
Lore is a local SQLite database managed by AI agents. You talk naturally — your data is structured, versioned, and stored on your machine.
Install
curl -fsSL https://getlore.sh/install | bash
This downloads the latest binary for your platform (macOS or Linux, x64 or arm64) to ~/.local/bin/lore.
You can customize the install with environment variables:
| Variable | Default | Description |
|---|---|---|
LORE_INSTALL_DIR | ~/.local/bin | Where to place the binary |
LORE_VERSION | latest | Pin a specific release tag |
Verify the installation:
lore --version
Initialize
lore init
This creates your database at ~/.lore/lore.db and installs AI skill files for your coding platforms.
You’ll be prompted to select which platforms to configure:
- Claude Code — writes skill files to
~/.claude/skills/lore/ - Cursor — writes a rule to
~/.cursor/rules/lore.mdc - Codex CLI — writes to
~/.agents/skills/lore/ - OpenCode — writes to
~/.agents/skills/lore/ - OpenClaw — writes to
~/.openclaw/workspace/skills/lore/
Or skip the prompt:
# Install for specific platforms
lore init --platforms claude,cursor
# Install for all platforms, accept all defaults
lore init --yes
# Skip skill file generation entirely
lore init --no-skills
First Use
Once initialized, your AI agent can start storing data immediately. The agent reads the current schema, creates tables as needed, and applies changes — all through structured JSON plans.
Here’s what a typical interaction looks like:
You say: “I had coffee with Sarah today, spent $4.50”
The agent:
- Runs
lore schemato check the current database state - Creates a plan to add a table (if needed) and insert a row
- Runs
lore apply -f -with the plan
You can read your data back anytime:
lore read --sql "SELECT * FROM expenses ORDER BY date DESC"
Or check what’s been stored:
lore status
What’s Next
- How It Works — understand the two-layer architecture
- CLI Reference — full list of commands and flags
- AI Integration — how skill files teach AI agents to use Lore