rw-paper-extractor
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill demonstrates a high standard of security and privacy by defaulting to local processing and adhering to explicit user-provided input for all operations.
- [COMMAND_EXECUTION]: The skill utilizes a Python script ('scripts/paper_case.py') to process PDF documents. Technical analysis of the script confirms it uses standard libraries such as PyMuPDF and Pillow to extract text and images locally. It does not perform any unauthorized command execution, system modifications, or shell injections.
- [EXTERNAL_DOWNLOADS]: The skill references established academic reporting guidelines from the EQUATOR Network, CONSORT, STROBE, and PRISMA. These are reputable, well-known academic resources used for verifying research methodology and do not involve the download of executable code or malicious payloads.
- [DATA_EXFILTRATION]: No data exfiltration logic was identified. The skill is designed to work on user-provided local files and explicitly states it will not bypass access restrictions or upload contents to external models without confirmation.
- [PROMPT_INJECTION]: The skill includes an ADHD-friendly output mode that is explicitly gated by user request. The implementation includes safety guards such as not storing diagnostic information and requiring confirmation for mode persistence, preventing potential privacy abuse.
Audit Metadata