start-feature

Warn

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill generates shell commands by directly interpolating the $ARGUMENTS variable into git commands without any escaping or validation. * Ingestion points: User-supplied input via the $ARGUMENTS placeholder in SKILL.md. * Boundary markers: None. No delimiters or safety instructions are used to restrict the interpretation of the user input as a literal string. * Capability inventory: Access to the Bash tool with a scope restricted to git commands. * Sanitization: None. Input is passed directly to the shell environment.
  • [COMMAND_EXECUTION]: The $ARGUMENTS variable is used to construct a branch name path in a git push command, creating a secondary injection point. * Evidence: git push -u origin feature/$ARGUMENTS
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 1, 2026, 04:50 AM
Security Audit — agent-trust-hub — start-feature