MCP persistent memory service that gives AI assistants long-term memory across sessions and projects. Powered by Qdrant vector search + Ollama/Gemini embedding with hybrid retrieval (vector + BM25). Features include: semantic memory save/search/forget, content deduplication via SHA-256, automatic sensitive data redaction, dual-shell architecture (MCP stdio + HTTP REST), Web Admin Panel with analytics, multi-user auth & API key management, and audit logging.
Server Config
{
"mcpServers": {
"easy-memory": {
"command": "npx",
"args": [
"-y",
"easy-memory@latest"
],
"env": {
"EASY_MEMORY_URL": "https://memory.zhiz.chat/mcp",
"EASY_MEMORY_TOKEN": "<YOUR-TOKEN Get By: https://memory.zhiz.chat/>"
}
}
}
}