Skip to content

Getting Started

Varun Pratap Bhardwaj edited this page Mar 17, 2026 · 2 revisions

Getting Started

Get SuperLocalMemory running in under 5 minutes.

Prerequisites

  • Python 3.11+ (python3 --version)
  • Node.js 14+ (only if installing via npm)
  • Any supported IDE (Claude Code, Cursor, VS Code, Windsurf, etc.)

Install

npm (recommended):

npm install -g superlocalmemory

pip:

pip install superlocalmemory

See Installation for git clone, platform-specific notes, and troubleshooting.

Setup

slm setup     # Choose mode A/B/C, configure provider
slm warmup    # Pre-download embedding model (~500MB, optional)

Modes:

  • A — Zero cloud. Data never leaves your machine. EU AI Act compliant. (default)
  • B — Local LLM via Ollama. Still fully private.
  • C — Cloud LLM for maximum accuracy (87.7% on LoCoMo).

Switch anytime: slm mode a, slm mode b, slm mode c.

Your First Memory

Store something:

slm remember "Our API uses JWT tokens with 24-hour expiry. Refresh tokens last 30 days."

Recall it later:

slm recall "JWT token expiry"

You should see the stored memory returned with a relevance score.

Verify Installation

slm status    # System info — mode, DB path, size
slm health    # Math layer health — Fisher-Rao, Sheaf, Langevin stats

Connect Your IDE

Auto-configure all detected IDEs:

slm connect        # Configure all detected IDEs
slm connect --list # See which IDEs are configured

Or add manually to your IDE's MCP config:

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

See IDE Setup for per-IDE instructions.

Web Dashboard

slm dashboard    # Opens at http://localhost:8765

17 tabs: memory browser, knowledge graph, recall lab, trust scores, math health, compliance, and more. Runs locally — no data leaves your machine.

Next Steps


Part of Qualixar | Created by Varun Pratap Bhardwaj

Clone this wiki locally