github-sensitive-data-cleanup

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses bundled Python scripts (scan_repo.py, rewrite_history.py, verify_cleanup.py, safe_push.py) to execute Git and GitHub CLI commands via the subprocess module. These commands are necessary for scanning, backing up, and rewriting repository history. The implementation uses argument lists rather than raw strings, which prevents standard command injection.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it ingests untrusted data from repository history during scanning and verification (primarily in scripts/scan_repo.py and scripts/verify_cleanup.py). The skill provides boundary markers in references/ai_semantic_review_prompt.md to scope the AI's review process. The skill's capability inventory includes file system access and shell command execution. Basic sanitization is performed in scripts/verify_cleanup.py by handling encoding errors during log retrieval.
  • [EXTERNAL_DOWNLOADS]: SKILL.md recommends installing reputable tools (gitleaks, git-filter-repo, gh) from well-known sources via the Homebrew package manager. These tools are standard for the skill's purpose and are not considered risky dependencies in this context.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 12:06 PM
Security Audit — agent-trust-hub — github-sensitive-data-cleanup