Skip to content
Hot eventDeveloping

Woolly v0.14.1 ships local memory engine with improved semantic search

1 reports1 reporting sourcesUpdated 20 h ago

Start here

AI digest

Woolly released v0.14.1 of its open-source memory engine, which runs locally by default and stores agent memories as readable Markdown files in ~/.woolly. The release improves semantic search accuracy, and with Ollama running the engine can also search by meaning, learn facts from chats and save decisions when a session ends. The project offers a mem0-shaped memory API with JavaScript and Python SDKs, MCP support for tools such as Claude Code, Cursor and Codex, and a woolly proxy that gives Ollama and OpenAI or Anthropic API apps the same memory. It is MIT licensed, requires no account or cloud, and keeps older versions of changed facts on disk. Woolly runs on the user's own machine and only answers that machine, so it fits agents run by an individual or team rather than a hosted service for thousands of users.

Generated from reports · updated 20 h ago

Report timeline

Follow the reports to see the event from different sides.

November 17
  1. Hacker News · AI
    Woolly launches local mem0-compatible memory API for AI agents

    Woolly released v0.14.1, an open-source memory engine that runs locally by default and stores agent memories as readable Markdown files in ~/.woolly. It offers a mem0-shaped memory API with JavaScript and Python SDKs, MCP support for tools such as Claude Code, Cursor and Codex, and a woolly proxy that gives Ollama and OpenAI or Anthropic API apps the same memory. The project is MIT licensed, requires no account or cloud, and keeps older versions of changed facts on disk.

Heat for this event

Current, comparable range
6
Peak, comparable range
10November 17 13:00
Change in 24 hours
–

The trend compares the same participants over hours observed continuously, so its range can be narrower than the current heat count. Hover or tap the chart for hourly heat; the arrow keys move through it.