Lore

A local-first memory layer for AI coding tools that preserves context across long sessions, preventing lossy summarization and re-explaining. It targets developers and engineering teams using AI assistants, storing memory in plain Markdown and an open SQLite database on the user's machine. It differentiates by avoiding proprietary vector databases, offering version-controlled team memory via Git and end-to-end encrypted sync.

Key features

• Gradient context
• Lore distillation
• Any provider support
• Recall tool
• .lore.md sync
• On-device vector search
• Import history
• Cost-aware caching
• Sessions lasting days
• Local-first storage
• End-to-end encrypted sync