Persistence

sjsam/Persistence

★ 0 stars Python 🤖 AI/LLM Updated today

Local, provider-agnostic MCP memory server — save an AI chat's context in one tool and recall it in another. SQLite + sqlite-vec with local Ollama embeddings.

View on GitHub → Try with Claude — $10 free →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "persistence": {
      "command": "uvx",
      "args": [
        "persistence"
      ]
    }
  }
}

Or install with pip: pip install persistence

README Excerpt

A local, provider-agnostic **memory server** for AI chats. Save the context of a conversation in one tool (Claude Desktop) and recall it in another (Google Antigravity, Ollama) without losing state — no more starting over when you switch providers or hit a rate limit. Every tool speaks **MCP** (Model Context Protocol), so Persistence is one local

Tools (8)

OLLAMA_URLPERSISTENCE_DATA_DIRPERSISTENCE_DB_PATHPERSISTENCE_EMBED_DIMPERSISTENCE_EMBED_MODELPERSISTENCE_HOSTPERSISTENCE_PORTPERSISTENCE_TOKEN

Topics

aiembeddingsfastmcpllmmcpmemorymodel-context-protocolollamaragsqlite-vec

Persistence

Quick Install

README Excerpt

Tools (8)

Topics

Related AI/LLM Servers