Memory and Code Participation

You run an MCP server setup that coordinates Claude Code, n8n workflow automation, and multiple MCP servers to manage memory, code generation targets, and filesystem interactions. This server enables AI-powered development processes, persistent project state, automated dumps, and structured code generation across frontend, backend, shared libraries, and documentation.

How to use

Interact with the MCP-enabled Claude Code assistant to manage project tasks, generate code into designated target directories, and perform memory dumps or file-system operations through dedicated MCP servers. You can issue natural language commands to Claude Code to add tasks, extract project insights, generate frontend or backend code, and trigger automated dumps. The included memory dump and filesystem servers expose capabilities for saving project state, organizing knowledge, and generating code into specific directories. Use Claude Code to drive actions such as creating todos, requesting a React component in the frontend folder, or exporting an n8n workflow to the workflows directory.

Typical workflows include: creating and updating tasks with priorities, generating code scaffolds in frontend/backend/shared target folders, requesting daily summaries and memory dumps, and running health checks. You can query for project status, upcoming actions, and recent learnings, then command the system to perform a dump or backup if needed. Remember to direct Claude Code toward the correct target directories when requesting code generation, and to verify that memory dumps and backups complete successfully after large changes or milestones.

How to install

Prerequisites you need before installation are: Node.js 18.0.0 or higher, npm 8.0.0 or higher, Git, and a Claude Code subscription capable of MCP integration. Windows users should prefer Git Bash or WSL2 for compatibility.

Follow these concrete steps to set up the MCP server suite and enable code generation and memory management through Claude Code:

1. Clone the project and enter the directory

git clone <your-repo-url>
cd my-project-n8n

2. Install dependencies

npm install

3. Run Windows-specific fixes if you are on Windows

./automation/windows-fix.sh

4. Set up target directories for code generation

setup-targets.sh

5. Configure environment variables for your MCP setup

# Copy example and edit
cp .env.example .env
nano .env

Configuration

MCP server configuration is defined to enable local, stdio-based servers that Claude Code can talk to directly. The following example shows how the two local MCP servers are configured to run on this setup.

{
  "mcpServers": {
    "memory-dump": {
      "command": "node",
      "args": ["./mcp-servers/memory-dump-server.js"],
      "env": {
        "MEMORY_FILE": "./project-memory.json",
        "DUMP_DIR": "./memory-dumps",
        "AUTO_DUMP": "true"
      }
    },
    "filesystem-target": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "${TARGET_PROJECT_DIR}"],
      "env": {}
    },
    "filesystem-current": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "."],
      "env": {}
    }
  }
}

Notes on environment and targets

Environment variables shown across the MCP configurations are included to ensure the servers receive the paths and flags they need. Common variables you may see include MEMORY_FILE, DUMP_DIR, and AUTO_DUMP for the memory dump server. Target directories are defined in the environment to direct generated code into frontend, backend, shared, and docs folders.

Troubleshooting

If MCP servers fail to connect, verify the status with Claude Code and restart as needed. Common issues include connectivity problems, permission errors, or Windows path translation needs.

Restart guidance and quick checks are:

• Check MCP status with the Claude Code prompt
• Restart Claude Code if necessary
• Confirm that target directories exist and have write permissions

Available tools

MemoryDump

Memory dump server features for saving project state, auto-dumps, and search analytics across memory content.

FilesystemServer

Filesystem MCP servers allow reading, writing, and managing files across target directories to structure frontend, backend, shared, and docs.

AutoDumpSystem

Automated memory preservation triggered by events and scheduled dumps to maintain project continuity.

SearchAnalytics

Enables searching and analyzing memory and project knowledge to surface decisions and learnings.