Cognitive AI Memory MCP Server

by sachitrafaio.github.sachitrafa/cognitive-ai-memoryv1.0.2

Ebbinghaus-based persistent memory for Claude. Memories decay with time, strengthen on recall.

270Dockerstdioofficial registry

context tax

queued

security

queued

cold start

queued

freshness

Maintained3mo ago

Install Cognitive AI Memory MCP server

claude mcp add cognitive-ai-memory -e DATABASE_URL='<database-url>' -- docker run -i --rm -e DATABASE_URL ghcr.io/your-memory/cognitive-ai-memory:1.0.2

{
  "mcpServers": {
    "cognitive-ai-memory": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "DATABASE_URL",
        "ghcr.io/your-memory/cognitive-ai-memory:1.0.2"
      ],
      "env": {
        "DATABASE_URL": "<database-url>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (project).

{
  "mcpServers": {
    "cognitive-ai-memory": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "DATABASE_URL",
        "ghcr.io/your-memory/cognitive-ai-memory:1.0.2"
      ],
      "env": {
        "DATABASE_URL": "<database-url>"
      }
    }
  }
}

Settings → Developer → Edit Config (claude_desktop_config.json), then restart.

{
  "servers": {
    "cognitive-ai-memory": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "DATABASE_URL",
        "ghcr.io/your-memory/cognitive-ai-memory:1.0.2"
      ],
      "env": {
        "DATABASE_URL": "<database-url>"
      }
    }
  }
}

Add to .vscode/mcp.json in your workspace.

{
  "mcpServers": {
    "cognitive-ai-memory": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "DATABASE_URL",
        "ghcr.io/your-memory/cognitive-ai-memory:1.0.2"
      ],
      "env": {
        "DATABASE_URL": "<database-url>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json.

Configuration

Variable	Required	Secret	Description
DATABASE_URL	yes	—	PostgreSQL connection string with pgvector. Example: postgresql://localhost:5432/yourmemory
OLLAMA_URL	—	—	Ollama server URL for local embeddings (nomic-embed-text) and classification (llama3.2:3b). Default: http://localhost:11434
EXTRACT_MODEL	—	—	Ollama model used for fact/assumption classification. Default: llama3.2:3b

Freshness

Maintained — last maintenance signal 3mo ago. The newest of the signals below sets the band.

Last commit (default branch)
2026-07-14 · 3mo ago · GitHub
Latest release
2026-05-25 · 5mo ago · GitHub · v1.4.26
Package published
no data · npm/PyPI
Registry entry updated
2026-03-23 · 7mo ago · official registry · v1.0.2

FAQ

›How do I install the Cognitive AI Memory MCP server in Claude Code?

Run: claude mcp add cognitive-ai-memory -e DATABASE_URL='<database-url>' -- docker run -i --rm -e DATABASE_URL ghcr.io/your-memory/cognitive-ai-memory:1.0.2. For Cursor, VS Code, Claude Desktop and Windsurf, use the install tabs above.

›Does Cognitive AI Memory require an API key?

Yes. It expects DATABASE_URL.

›Can I use Cognitive AI Memory as a remote (hosted) MCP server?

No hosted endpoint is published; it runs locally over stdio.

›Is Cognitive AI Memory in the official MCP registry?

Yes, as io.github.sachitrafa/cognitive-ai-memory.