Open source context drive for all your AI agents | Puppyone

Puppyone is an open-source context drive that connects diverse SaaS data sources to AI agents through a unified, versioned file system. It provides Git-style history tracking for all agent actions and granular permission scoping to ensure secure data access. By offering multiple access methods like MCP and REST APIs, it serves as a robust infrastructure for managing multi-agent collaboration.

Key Points

• Unified File System: Ingests data from multiple SaaS sources and structures it into agent-readable formats like JSON and Markdown.
• Git-Style Versioning: Automatically tracks every modification made by agents, providing a full audit trail and the ability to roll back to any previous state.
• Scoped Access Control: Enables granular permissions, allowing users to restrict specific agents to specific directories or files.
• Multi-Interface Compatibility: Supports various connection methods including MCP for Cursor/Claude, a CLI for terminal use, and a REST API for automation.
• Open Source and Self-Hostable: Offers a Docker-based deployment option that keeps all metadata and files within a user's own infrastructure.