repo-classify
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill (SKILL.md) documents repository management procedures and identifies sensitive data paths (e.g., .env, financial data, legal documents) that must be excluded from version control via .gitignore. It also references local configuration storage in the user's home directory (~/.clawfirm/skills/), which is standard for local tool persistent storage.- [PROMPT_INJECTION]: The skill (SKILL.md) has an indirect prompt injection surface as it processes external files for classification. If these files contain malicious instructions, the agent could potentially follow them during processing. Ingestion points: User-provided repository files (referenced in SKILL.md). Boundary markers: Absent. Capability inventory: Subprocess calls for git operations (add, commit, push) and file-write via cp (SKILL.md). Sanitization: Absent.
Audit Metadata