git-flow-master

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Git logs and diffs, representing a theoretical injection surface.\n
  • Ingestion points: Silently reads repo state using git status, git branch, git diff, and git log (SKILL.md, Step 1).\n
  • Boundary markers: The skill does not explicitly use delimiters or instruction-shielding for the data read from shell commands.\n
  • Capability inventory: Shell command execution (git, gh, bun) and file system writes (commits, temporary files).\n
  • Sanitization: Employs temporary files for PR bodies to prevent command-injection via shell escaping.\n- [COMMAND_EXECUTION]: The skill automates version control via shell commands.\n
  • Evidence: Uses git for repository management and gh for pull request creation.\n
  • Evidence: Runs project-specific policy reconciliation via bun run git:policy.\n- [SAFE]: The skill provides explicit guidance against insecure practices like force-pushing to shared branches or committing environment files and secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 08:14 PM
Security Audit — agent-trust-hub — git-flow-master