pdf-ppt
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill contains a file named
pdf-ppt_audit_result_v1.jsonwhich provides self-referential safety claims, such as "security: PASS" and "grade: Production Ready". This is a known pattern (Category 8e) where a skill includes data designed to influence the findings of an automated security analyzer. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted research paper PDFs to generate summaries and slide content, creating a vulnerability surface for indirect prompt injection.
- Ingestion points: External PDF files and figure legends extracted during the parsing workflow.
- Boundary markers: Absent. The instructions do not define clear delimiters or specify that the agent should ignore instructions embedded within the parsed research papers.
- Capability inventory: The skill uses local libraries (PyMuPDF, pdfplumber) and tools (GROBID) to read from and write to the local file system (e.g.,
D:\\SKILL\\project\\PPTX\\SKILL.md). - Sanitization: No evidence of sanitization or filtering of the extracted PDF text before it is used to generate Chinese summaries and interpretations.
Audit Metadata