sandbox-agent
Pass
Audited by Gen Agent Trust Hub on Mar 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEREMOTE_CODE_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download installation scripts and binaries from the vendor's release domain (releases.rivet.dev) and well-known technology services (e.g., Docker, Go, Rustup). These are standard setup procedures for the orchestration server.
- [REMOTE_CODE_EXECUTION]: Employs piped shell execution (curl | sh) for software installation and provides a managed process API to run code within sandboxes, which is the primary intended functionality of the toolkit.
- [COMMAND_EXECUTION]: Includes documentation for system-level commands, package management, and desktop environment configuration required for the operation of the coding agents.
- [CREDENTIALS_UNSAFE]: Features a built-in utility (credentials extract) designed to help developers retrieve and migrate their local AI provider API keys into the sandbox environment.
Audit Metadata