remote-run

Pass

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes instructions to install the bb CLI utility using curl -fsSL https://install.buildbuddy.io | bash. This fetches and executes an installation script directly from the vendor's official domain.
  • [COMMAND_EXECUTION]: The skill facilitates the execution of Bazel and arbitrary shell commands on remote infrastructure via the bb remote command. This includes the use of the --script flag for executing shell scripts, which is a core feature of the tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill manages a local configuration file (persistent-flags.md) to store command-line flags. While this creates a surface for data ingestion, the skill requires explicit human confirmation and syntax verification before adding new flags, providing a human-in-the-loop mitigation.
  • Ingestion points: Data is read from persistent-flags.md and interpolated into bb remote commands.
  • Boundary markers: None identified for the file content itself.
  • Capability inventory: The skill can execute shell commands through the bb CLI.
  • Sanitization: The instructions mandate user confirmation of the flag syntax and explicit permission before the agent modifies the persistent flags file.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 12, 2026, 07:15 PM
Security Audit — agent-trust-hub — remote-run