Index project files and conversations into the MemPalace memory system.
npx skills add https://github.com/dimetron/pi-go --skill memory-mine
Please help me install this Agent Skill. Skill: memory-mine Source: https://github.com/dimetron/pi-go/tree/main/.pi-go/skills/memory-mine Command: npx skills add https://github.com/dimetron/pi-go --skill memory-mine
Mine project files or conversations into the MemPalace memory system. Indexes source code and conversation history as searchable palace drawers with semantic embeddings and room assignment.
High-intent search queries and answers about installing and using this skill.
To index source code and conversation history into a searchable knowledge base, this Skill ingests project files and chats, generates semantic embeddings, assigns content to memory drawers, and deduplicates entries.
Deduplicating conversation history and code files using embeddings is handled automatically during ingestion, organizing content into searchable memory drawers while applying semantic embeddings to identify and remove redundant data.
Yes, you need to initialize palace.db before mining project files into memory. An available embedding model is also required to generate semantic vectors for the code and conversation content.
You can organize indexed source code into specific rooms by configuring assignments through a mempalace.yaml file or by relying on the existing directory structure to automatically route drawers.
To specify conversation history ingestion separately from code mining, execute the command with the --convos parameter, and use the optional --wing parameter to target specific sections of the memory palace.
patricio0312revConfigure Pinecone, Chroma, pgvector, and Qdrant vector databases for semantic search.
zaebeeStore and retrieve negotiation history and facts using PostgreSQL with pgvector.
robwestzMonitor and report MemPalace knowledge graph operational status metrics.
MSamiulHasnatIndex embeddings and perform context-aware similarity queries for document retrieval.
