Introduction

Lore is an open-source coordination memory layer for multi-agent AI systems. When several agents work on one task, Lore gives them a single shared memory with a read-your-writes guarantee: when one agent writes, any later context request reflects that write — as a distilled memory, or as a raw tail until extraction catches up.

What it does

• Write events to a run; Lore extracts and consolidates them into memories and claims.
• Pack a token-budgeted context for a query, with min_seq so the pack is guaranteed to reflect what you just wrote.
• Inspect the memories, their version history, and each run's pack history.

Run it

Lore self-hosts in a single docker compose up — no account, no cloud. The default extractor is an offline, deterministic fixture, so the whole write → consolidate → pack loop runs with no API key.

Next steps