conventional-git

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to generate Git commands and commit messages based on user-supplied descriptions. While the skill enforces strict formatting for branch names (lowercase, hyphens only), which provides inherent sanitization, it does not explicitly instruct the agent to use boundary markers or delimiters when interpolating untrusted user input into commit bodies or extended descriptions.
  • Ingestion points: User prompts requesting branch creation, worktree setup, or commit message generation.
  • Boundary markers: Absent in instructions regarding user input interpolation.
  • Capability inventory: Access to git and gh (GitHub CLI) via the Bash tool.
  • Sanitization: The skill's naming rules (lowercase, no special characters) serve as a partial sanitization mechanism for branch and worktree names, but no explicit sanitization is provided for commit messages.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:59 PM
Security Audit — agent-trust-hub — conventional-git