pydantic-ai-backend
View on GitHubFile Storage & Sandbox Backends for Pydantic AI: console tools for file operations, Docker-isolated sandboxes for safe execution, and permission system with presets for access control. Enables secure multi-user handling and testing in agents via in-memory, local, or containerized storage.
Adds workspaces, file and shell tools, and permission controls to Pydantic AI agents. Supports Docker, Kubernetes, Daytona, remote sandboxd, and in-memory backends for isolated code execution and persistent agent sessions.
Use Cases
Run agent-written code in isolated sandboxesBuild AI coding assistants with file and shell toolsCreate read-only code review agentsProvide persistent per-session workspaces for multi-user agentsTest agent workflows with in-memory filesystems
Built With
- Language
- Python
- Frameworks
- Pydantic AI · FastAPI
Tags
Pydantic AI · agent workspaces · sandbox · code execution · console tools · permissions · Docker · Kubernetes · multi-user agents