Getting Started
Set up Lore in under 5 minutes. Zero configuration required.
What is Lore?
Lore is persistent, project-level memory for developers. It captures the why behind your
code — architectural decisions, invariants, gotchas, and abandoned approaches — and stores them as
structured JSON alongside your codebase in a .lore/ directory.
Every entry is versioned in git, so your team automatically shares the project's institutional knowledge. When you use AI coding tools like Claude or Cursor, Lore injects this context directly into your sessions via MCP.
Quick Start
Get Lore running in your project in 4 steps:
1. Install globally via npm
$ npm install -g lore-memory
2. Initialize in your project
$ cd your-project
$ lore init
This creates a .lore/ directory in your project root, installs a git post-commit hook, and
scans for existing code comments.
3. Start capturing context
$ lore watch --daemon
The watcher runs in the background, scanning for signal phrases like WARNING:,
HACK:, and IMPORTANT: in your code comments and automatically drafting Lore
entries.
4. Generate an AI prompt
$ lore prompt "refactoring auth" | pbcopy
Lore uses semantic vector search to find the most relevant entries for your query and compiles a formatted system prompt ready for any LLM.
lore with no arguments to launch the interactive menu.
What's Next?
- Installation — prerequisites, alternative install methods
- Entry Types — understand the 4 knowledge categories
- CLI Reference — all 16 commands with options
- Integrations — set up Claude Code or Cursor via MCP