compact-dir-listing
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides instructional content and standard Python code for repository exploration and token optimization. It promotes efficiency and token reduction without introducing malicious functionality.
- [INDIRECT_PROMPT_INJECTION]: The skill describes methods for exploring file system structures which may include untrusted data in the form of filenames.
- Ingestion points: The skill uses git ls-files, fd, and a Python script using pathlib.Path.rglob to read filenames from the local file system.
- Boundary markers: No explicit boundary markers or warnings against embedded instructions in filenames are provided in the output formatting.
- Capability inventory: The skill's Python code performs read-only file system traversal using pathlib.
- Sanitization: No sanitization or escaping of filenames is implemented in the provided Python utility.
Audit Metadata