LoreKitPersistent memory
for your AI agents
Your AI agent forgets everything when a session ends. LoreKit gives it persistent memory — so it never asks the same question twice.
Sign in free · 2-minute setup
Watch memories form and persist
Session A writes. Session B loads. No re-explaining.
memory store
— none yet —
Memories survive session ends · Shared across Claude Code, Cursor & Codex
Get started in 2 minutes
Free · Open source
- 1
Sign in and get your token
GitHub, email + password, or a magic link — takes 30 seconds
- 2
Run one command
npx @lorekit/cli install - 3
Your agent now remembers
Memories persist across sessions, repos, and tools — automatically
Agents that remember
LoreKit gives your AI coding agents a shared memory store. Memories written in one session are available in the next — across repos, branches, and tools.
Scope-aware memory
Memories are namespaced by scope: global, project, repo, or branch. Agents read the narrowest scope first and fall back automatically.
Wired to your workflow
Connect via MCP in one config line. A GitHub webhook turns resolved PR review comments into candidate memories automatically.
Browse and curate
The dashboard lets you explore every memory your agents have written, archive stale entries, and watch activity in real time.