More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    mem20 install

    1

    Install and configure mem20 or mem20-ce as a persistent MCP memory server for AI agents.

    $3.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    mem20 install

    mem20 install

    Example session with this skill installed

    I need to install mem20-ce for my Cursor setup and make sure it saves to a local folder.

    • Read your context and instructions
    • Compiled the mem20 install

    I will install mem20-ce and generate the MCP configuration for Cursor.

    Installation and Configuration

    First, install the Core Edition:

    python3 -m pip install --upgrade pip
    pip install mem20-ce
    

    Next, add this configuration to your Cursor MCP settings. I have expanded the store path to be absolute to ensure stability:

    {
      "mcpServers": {
        "mem20": {
          "command": "mem20-mcp",
          "env": {
            "MEM20_STORE_PATH": "/Users/username/.mem20-ce/store"
          }
        }
      }
    }
    

    Verification

    To verify the installation, run these checks:

    1. Check Health:
      curl -s localhost:8080/health
      

    Test Memory: Ask me to "remember that the project deadline is Friday" and then ask "what is the project deadline?" to verify the round-trip.

    Next steps

    • Restart Cursor to load the new MCP configuration.
    • Confirm mem20 appears in your tool list.
    • Run a contamination check to ensure a clean store.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Setting up persistent memory for AI agents often involves complex environment variables and configuration mismatches. This skill handles the installation and wiring of mem20 and mem20-ce as an MCP memory server for clients like Claude Code and Cursor.

    What it does

    • Environment installation handles pip installation for mem20-ce or full monorepo setup via venv and requirements.
    • MCP configuration generates valid JSON config blocks for server commands, arguments, and environment variables.
    • Store management configures stable MEM20_STORE_PATH settings to ensure memory persistence across sessions.
    • Service verification performs health checks and remember-recall round-trips to confirm the server is responsive.
    • Agent integration provides specific wiring instructions for Claude Code, Cursor, Codex, and Hermes.

    How it works

    1. Choose edition by selecting between the lightweight mem20-ce or the full tool-heavy mem20 monorepo.
    2. Execute install using the provided pip or git workflows to build the environment.
    3. Configure client by mapping the server entrypoint and store path into the agent's MCP settings.
    4. Verify health via curl commands to the health and ready endpoints.

    Frameworks & tools

    This skill works with Python 3, MCP (Model Context Protocol), pip, and git. It supports clients including Claude Code, Cursor, Codex, and Hermes.

    Why this beats prompting it yourself

    Manual MCP setup often fails due to relative path issues in environment variables or incorrect entrypoint mapping. This skill provides precise, absolute path configurations and verification steps that ensure the memory store is actually writing to disk before you start using it.

    Use cases

    • Add persistent memory to a new Claude Code or Cursor project.
    • Migrate from a standard agent to a mem20-grounded memory architecture.
    • Troubleshoot broken MCP connections between an agent and its memory store.

    Known limitations

    Requires an existing Python 3 environment. Does not perform ongoing memory audits or deep memory model explanations.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.

    Does not ship a prebuilt binary — installs from PyPI (mem20-ce) or GitHub (full monorepo). MCP config paths differ by agent; skill covers major hosts but you may need to adapt JSON. Does not migrate existing stores between paths automatically.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 8 days ago

    Software
    Python 3.10+mem20-ceGit
    Optional
    mem20 (full monorepo

    Permissions

    Network Access
    Environment Variables
    Write Files
    Terminal / Shell
    Read Files

    File Scopes

    ~/.mem20/**
    ~/.mem20-ce/**
    **/mcp*.json
    **/.cursor/**/mcp*.json

    Shell and network for pip/git install. Write access for MCP config and store path. Env vars for MEM20_STORE_PATH.

    Listed8 days ago

    What's inside

    Frequently Asked Questions

    Only if you want the deeper wiring guide, troubleshooting, and agent-specific MCP configs. Harness offers a minimal path; this skill is the full install playbook.

    Start with mem20-ce (pip install mem20-ce) unless you need the large subsystem tool surface.

    Any MCP client — Claude Code, Cursor, Codex CLI, Hermes, and similar.

    New in LLM Ops