Index daily chat logs and enable semantic search across past conversations.
npx skills add https://github.com/cutec-chris/PawLia --skill memory-cutec-chris
Please help me install this Agent Skill. Skill: memory Source: https://github.com/cutec-chris/PawLia/tree/main/skills/memory Command: npx skills add https://github.com/cutec-chris/PawLia --skill memory-cutec-chris
Long-term memory across conversations with persistent indexing and semantic search to recall past discussions. Use when the user asks about something from past conversations, wants to recall what was discussed days/weeks ago, or says things like "we talked about...", "remember when...", "what did I say about...". The query should be a natural language question about past conversations. Special commands: "index" to manually trigger indexing of new chat logs, "status" to show indexing status.
Use the memory skill to search past conversations or manually trigger indexing and status checks.
High-intent search queries and answers about installing and using this skill.
To enable long-term memory for semantic search across past chat logs, you need a configured embedding provider, embedding model, embedding host, and an optional rag backend. Indexing daily logs allows persistent recall of prior discussions across sessions.
The best way to recall previous conversations is using semantic search over indexed daily chat logs. Query with natural language questions like "what did I say about" to retrieve context from past sessions without manual scrolling.
Yes, this RAG memory skill requires a configured embedding provider, embedding model, and embedding host to perform semantic search. An optional rag backend provides retrieval with proper context and latency controls.
Yes, you can manually trigger indexing of new chat logs by using the "index" command. The "status" command shows current indexing status to verify that memory is up-to-date across sessions.
Semantic search for conversation recall is not suited for queries outside past chat history or when lacking a configured embedding provider. It only retrieves indexed daily logs and cannot access unindexed or external data sources.
CleanExpoGenerate pgvector embeddings and run similarity or hybrid search in Supabase/PostgreSQL.
7xuanluSearch Wenlan's local memory by natural-language query.
466852675Configure persistent memory search in agent JSON settings.
rukikotooQuery the econometrics RAG knowledge base to retrieve relevant document chunks.
