Persist and semantically search learnings across AI agents.
npx skills add https://github.com/MemoAiTools/skills --skill memoai
Please help me install this Agent Skill. Skill: memoai Source: https://github.com/MemoAiTools/skills/tree/main/memoai Command: npx skills add https://github.com/MemoAiTools/skills --skill memoai
MemoAI provides persistent memory across coding sessions, enabling search and recall of past bug fixes, architectural decisions, implementation patterns, and best practices to accelerate debugging and development.
Ask memoai_memo_search for relevant learnings, or memoai_memo_record to capture new insights.
High-intent search queries and answers about installing and using this skill.
You can persist and semantically search AI agent learnings by storing memos with embeddings, enabling recall of past bug fixes and architectural decisions to reduce rework across sessions.
Persistent memory for AI agents stores non-trivial solutions and implementation patterns for future reuse. You need it when debugging or developing to surface relevant past learnings before coding.
To record new architectural decisions, use memoai_memo_record to capture insights, and use memoai_memo_search to semantically search past bug fixes and implementation patterns for cross-session knowledge sharing.
Yes, persistent memory for AI agents requires PostgreSQL to store memos and ChromaDB for embeddings. You also need to set the MEMOAI_PROJECT environment variable and use an MCP client.
Yes, the persistent memory knowledge management system uses an async Gardener classifier to reclassify episodes automatically, organizing stored learnings for more accurate semantic search retrieval.
Limitations include requiring a specific MEMOAI_PROJECT environment variable and MCP client dependency. Without PostgreSQL and ChromaDB configured, semantic search and cross-session knowledge sharing cannot function.
ever-oliStore and query embeddings and metadata with the chromadb vector database.
opencodosLocate past conversations and decisions in an Obsidian Vault using hybrid search.
jonkikyConsolidate and retrieve project learnings across sessions.
zangxin75Implement a three-layer persistent memory system using Markdown, ChromaDB, and NetworkX.
