superlocalmemory

Overview

SuperLocalMemory is a privacy-focused AI memory system that operates entirely on your device, sending no data to external APIs or cloud services. It implements advanced mathematical techniques from differential geometry, algebraic topology, and stochastic analysis to provide memory retrieval without requiring external LLM calls. The system offers multiple modes: Mode A (local-only, zero-LLM), Mode B (with optional cloud), and Mode C (with local LLM). It provides dual interfaces as both an MCP server and a CLI tool, with extensive documentation and IDE integrations.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• forget

  Programmatic memory archival via lifecycle rules

  Try:Archive memories older than 30 daysTry:Forget all memories about project X
• quantize

  Trigger smart compression on demand

  Try:Compress low-priority memoriesTry:Run memory quantization
• consolidate_cognitive

  Extract and store patterns from memory clusters

  Try:Extract patterns from customer feedback memoriesTry:Consolidate technical documentation insights
• get_soft_prompts

  Retrieve auto-learned patterns for context injection

  Try:Show me the learned patternsTry:What soft prompts are available?
• reap_processes

  Clean orphaned SLM processes

  Try:Clean up zombie processesTry:Reap abandoned memory processes
• get_retention_stats

  Memory lifecycle analytics

  Try:Show memory retention statisticsTry:How strong are my memories?
• remember

  Store new information in memory

  Try:Remember that Alice works at GoogleTry:Store this important fact: meeting at 2pm tomorrow
• recall

  Retrieve information from memory

  Try:What does Alice do?Try:Find information about the database schema
• status

  Check memory system status

  Try:Show memory system statusTry:How is my memory performing?
• warmup

  Pre-download embedding models

  Try:Pre-download embedding modelsTry:Warm up the memory system
• migrate

  Migrate memory data between versions

  Try:Migrate memories from V2 to V3Try:Upgrade memory data
• doctor

  Verify memory system is working correctly

  Try:Check system healthTry:Run diagnostic tests

Comparable tools

Installation

npm install -g superlocalmemory
slm setup     # Choose mode (A/B/C)
slm doctor    # Verify everything is working
slm warmup    # Pre-download embedding model (~500MB, optional)

pip install superlocalmemory

{
  "mcpServers": {
    "superlocalmemory": {
      "command": "slm",
      "args": ["mcp"]
    }
  }
}

FAQ

How does SuperLocalMemory work without cloud LLMs?

It uses advanced mathematical techniques from differential geometry, algebraic topology, and stochastic analysis to replace the work typically done by LLMs for similarity scoring, contradiction detection, and lifecycle management.

Which AI clients support SuperLocalMemory?

It works with any MCP-compatible client including Claude Code, Cursor, Windsurf, VS Code Copilot, Continue, Cody, ChatGPT Desktop, Gemini CLI, JetBrains, Zed, and Antigravity.

Compare superlocalmemory with