daytona-dev

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the Daytona CLI from GitHub releases, a well-known and trusted software distribution platform.
  • [COMMAND_EXECUTION]: Orchestrates the development environment using local scripts located in the .devcontainer/ directory and the daytona CLI tool for sandbox lifecycle management, which is expected behavior for a development environment skill.
  • [PROMPT_INJECTION]: The skill includes instructions to execute a script that accepts a user-provided branch or commit identifier as an argument. This provides an ingestion point for external data, which is a standard feature of development tools.
  • Ingestion points: Branch or commit argument in SKILL.md used by the test-on-daytona.sh script.
  • Boundary markers: Not explicitly defined in the instructions.
  • Capability inventory: Uses bash and daytona exec to run commands in the cloud sandbox.
  • Sanitization: Not observed in the skill instructions; relies on the underlying shell and script logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 12:27 PM
Security Audit — agent-trust-hub — daytona-dev