Penfield

Persistent memory for AI agents. Store decisions, preferences, and context that survive across sessions. Build knowledge graphs that compound over time. Works with Claude, Cursor, Windsurf, Gemini CLI, and any MCP-compatible tool.

Quick Start

Claude (Desktop, Mobile, Web)

Add as a custom connector in Settings → Connectors:

Name: Penfield
Remote MCP server URL: https://mcp.penfield.app

Claude Code

claude mcp add --transport http penfield https://mcp.penfield.app

Cursor

One-click install:

Install Penfield in Cursor

Or add manually to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "Penfield": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.penfield.app/"]
    }
  }
}

Windsurf, Cline, Roo Code, and Others

Add to your MCP configuration file:

{
  "mcpServers": {
    "Penfield": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.penfield.app/"]
    }
  }
}

App	Config Location
Windsurf	`~/.codeium/windsurf/mcp_config.json`
Cline	VS Code Settings → Cline → MCP Servers
Roo Code	VS Code Settings → Roo Code → MCP Servers
Zed	`~/.config/zed/settings.json` under `"context_servers"`

Gemini CLI

gemini mcp add penfield -- npx -y mcp-remote https://mcp.penfield.app/

Or add to ~/.gemini/settings.json:

{
  "mcpServers": {
    "Penfield": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.penfield.app/"]
    }
  }
}

What You Get

17 tools for persistent memory:

Category	Tools
Memory	`store`, `recall`, `search`, `fetch`, `update_memory`
Knowledge Graph	`connect`, `disconnect`, `explore`
Context	`awaken`, `reflect`, `save_context`, `restore_context`, `list_contexts`
Artifacts	`save_artifact`, `retrieve_artifact`, `list_artifacts`, `delete_artifact`

Hybrid search combining BM25 (keyword), vector (semantic), and graph (connections) for recall that actually finds what you need.

Cross-platform sync — same memory, same knowledge graph, regardless of which tool you connect from.

How It Works

Sign up at portal.penfield.app/sign-up
Connect using one of the methods above
Authenticate when prompted (OAuth flow)
Start using — your agent now has persistent memory

Every session should start with:

awaken()     # Load identity and personality context
reflect()    # Orient on recent work (default: last 7 days)

Without these, your agent starts cold with no context.

Documentation

Tools Reference — All 17 tools with parameters and examples
Memory Types — The 11 memory types and when to use each
Relationships — The 24 relationship types for connecting memories
AI Agent Guide — Instructions for AI agents using Penfield

Use Cases

Personal assistant that remembers

Your preferences compound over time
Picks up conversations where you left off
Learns how you like things done

Development workflows

Track investigation threads across sessions
Remember architectural decisions and why they were made
Hand off context between coding sessions

Research and writing

Build knowledge graphs of connected ideas
Store insights and corrections as understanding evolves
Checkpoint progress on long-running projects

Also Available

OpenClaw Native Plugin — If you use OpenClaw, the native plugin is 4-5x faster (no MCP proxy layer):

openclaw plugins install openclaw-penfield
openclaw penfield login

openclaw-penfield on GitHub · openclaw-penfield on npm

API — Direct HTTP access at api.penfield.app for custom integrations.

Penfield Memory