Show HN: Memweave CLI – search your AI agent's memory from the shell

Category: library

Tags: ai-agents, vector-search, memory-management, python-library, sqlite, cli-tool, markdown, hybrid-search

Score: 7.5/10 (Innovation: 7, Technical: 8, Documentation: 8, Utility: 7)

Memweave provides AI agents with persistent, searchable memory stored as plain Markdown files and indexed by SQLite, combining BM25 keyword search with semantic vector search. Its innovative approach of using human-readable files as the source of truth with a rebuildable SQLite index offers a unique blend of transparency, offline capability, and zero-infrastructure agent memory. The project is well-documented with clear concepts, usage examples, and a comprehensive API reference.

Target audience: AI/ML engineers, backend devs, agent developers

Repository: https://github.com/sachinsharma9780/memweave · Python · MIT · 27 stars

View on Hacker News