Unify conversations, memories, and file references into a local searchable store.
npx skills add https://github.com/wipcomputer/memory-crystal --skill wip-memory-crystal
Please help me install this Agent Skill. Skill: wip-memory-crystal Source: https://github.com/wipcomputer/memory-crystal/tree/main Command: npx skills add https://github.com/wipcomputer/memory-crystal --skill wip-memory-crystal
Memory Crystal unifies the memories of multiple AI agents by storing conversations, memories, and file references in a private, local store, enabling cross-agent recall and efficient searching.
Use the memory-crystal skill to recall a detail from a past conversation or search for a topic across all your AIs.
High-intent search queries and answers about installing and using this skill.
You can share memory across multiple AI agents by storing conversations, facts, and file references in a unified local store, enabling cross-agent recall and search.
Yes, the shared memory store works with Claude Code CLI, OpenClaw, and MCP-compatible clients, unifying their conversations and memories into a single source of truth.
Local memory search uses a hybrid indexing approach combining BM25 and vector search to retrieve past discussions, facts, and preferences from a private, offline store.
The best way to manage remembering and forgetting facts is using a sovereign local store that acts as a single source of truth for all connected AI agents.
Yes, multi-device sync is supported via a secure relay, allowing your private local memory store to stay synchronized across different devices.
You recall past conversation details by running a cross-AI search against the unified memory store, which surfaces relevant discussions, notes, and shared context.
Human-Frontier-Labs-IncPersist session memory and retrieve contextual patterns with AgentDB.
bmsull560Integrate vector-based memory with knowledge graphs for multi-session conversation context.
kennykankushCapture conversation essence into the active host's memory directory.
ai-kitchen-incChunk documents, generate embeddings, and index vectors in Pinecone for RAG.
