# Footprint

> Footprint is an open-source MCP (Model Context Protocol) server that automatically captures and encrypts AI conversations as verifiable records. It provides end-to-end encryption, Git timestamps, and tamper-proof exports for developers, researchers, and content creators who need audit trails for their LLM interactions.

## Key Information

- **Package**: `@pcircle/footprint` on npm
- **License**: MIT (free for personal and commercial use)
- **Platform**: Cross-platform (Node.js 20+)
- **Primary Use Case**: Audit trail and evidence collection for AI/LLM conversations
- **Integration**: Model Context Protocol (MCP) for Claude Desktop and other MCP-compatible clients

## Core Features

- **Automatic Capture**: Intercepts LLM conversations via MCP protocol without manual logging
- **End-to-End Encryption**: XChaCha20-Poly1305 encryption with Argon2id key derivation
- **Git Timestamps**: Cryptographic proof of when conversations occurred
- **Tamper-Proof Export**: ZIP archives with SHA-256 checksums
- **Smart Tagging**: Organize and search conversations with tags
- **100% Local**: Data never leaves your machine

## Installation

```bash
npm install -g @pcircle/footprint
```

## Configuration

Add to Claude Desktop config (`~/Library/Application Support/Claude/claude_desktop_config.json`):

```json
{
  "mcpServers": {
    "footprint": {
      "command": "npx",
      "args": ["@pcircle/footprint"],
      "env": {
        "FOOTPRINT_DB_PATH": "/path/to/evidence.db",
        "FOOTPRINT_PASSWORD": "your-secure-password"
      }
    }
  }
}
```

## MCP Tools

- `capture-footprint`: Capture and encrypt a conversation
- `list-footprints`: List all captured evidence with pagination
- `get-footprint`: Retrieve and decrypt specific evidence
- `export-footprints`: Export to encrypted ZIP archive
- `search-footprints`: Search through captured conversations
- `verify-footprint`: Verify evidence integrity

## Documentation

- [GitHub Repository](https://github.com/PCIRCLE-AI/footprint-mcp)
- [npm Package](https://www.npmjs.com/package/@pcircle/footprint)
- [Website](https://footprint.memesh.ai)

## Use Cases

1. **IP Protection**: Timestamp your AI-assisted creative work
2. **Compliance**: Maintain audit trails for regulated industries
3. **Research**: Track and verify AI-generated research contributions
4. **Legal**: Create verifiable evidence of AI interactions
5. **Development**: Document AI-assisted code generation

## Security

- Encryption: XChaCha20-Poly1305 (256-bit)
- Key Derivation: Argon2id (OWASP recommended)
- Storage: Local SQLite with encrypted BLOBs
- No cloud, no tracking, no data collection

## Pro Features (Coming Soon)

- Cloud sync across devices
- Blockchain timestamp verification
- AI-powered summaries
- Semantic search
- Multi-device support

## Company

- **Developer**: PCIRCLE, LLC
- **Website**: https://footprint.memesh.ai
- **Support**: https://github.com/PCIRCLE-AI/footprint-mcp/issues