skills/tristan-mcinnis/ppt-translator-formatting-intact-with-llms/ppt-translator/Gen Agent Trust Hub
ppt-translator
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted PowerPoint files and passes their text content directly to LLM providers for translation, creating a surface for indirect prompt injection.
- Ingestion points: Text extracted from slides in
scripts/ppt_translator/pipeline.pyviaextract_text_from_slide. - Boundary markers: Missing. The text content is interpolated directly into system instructions in
scripts/ppt_translator/providers/base.py,anthropic_provider.py, andgemini_provider.pywithout delimiters or 'ignore embedded instructions' warnings. - Capability inventory: The skill has capabilities to read/write local files (PPTX and XML) and perform network operations to LLM APIs (OpenAI, Anthropic, DeepSeek, etc.).
- Sanitization: No filtering or escaping is performed on the extracted slide content before it is sent to the LLM.
Audit Metadata