skills/buiducnhat/cobrew/git-commit/Gen Agent Trust Hub

git-commit

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill includes explicit security instructions to never commit secrets, private keys, or .env files, following security best practices.
  • [COMMAND_EXECUTION]: The skill utilizes git and bash commands for staging files and committing changes, which are standard for version control tasks.
  • [SAFE]: Constraints are provided to prevent destructive git actions such as force pushes to main branches or unauthorized configuration changes.
  • [PROMPT_INJECTION]: The skill processes untrusted output from git diff to generate commit messages, representing an indirect prompt injection surface (Category 8). \n
  • Ingestion points: git diff and git status output (SKILL.md). \n
  • Boundary markers: Absent. \n
  • Capability inventory: git add, git commit, and bash tool execution (SKILL.md). \n
  • Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 06:37 AM
Security Audit — agent-trust-hub — git-commit