Tag
This paper presents a reusable template (llm-wiki-memory-template) that implements the llm-wiki pattern to preserve session memory and failure paths in collaborative knowledge work involving multiple humans, AI agents, and domains. It argues for the substrate's generalizability across three axes and includes case studies demonstrating improved evidence tracking.
LangChain launches OpenWiki Brains, a framework for proactive memory for AI agents that automatically builds and updates wikis from connected sources like Gmail, Notion, and git repos.
OpenWiki 0.1.0 introduces OpenWiki Brains, a tool that creates and maintains a local wiki from sources like Gmail, Notion, git repos, and web search, giving AI agents proactive, durable memory without manual updates.
The article discusses the emerging pattern of 'wiki memory' for AI agents, where raw source data is intelligently compressed into a persistent, structured knowledge layer that agents can use efficiently. It compares this to basic RAG and gives examples like DeepWiki and LLM Wiki.