The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the ingestion and processing of untrusted external scientific papers which may contain malicious instructions.
Ingestion points: The skill accepts paper PDFs, supplementary figures, and structured reading notes as input sources (SKILL.md, 'Accepted Inputs' section).
Boundary markers: The instructions do not define delimiters or provide specific warnings to the agent to ignore or isolate instructions potentially embedded within the source materials.
Capability inventory: The skill utilizes PyMuPDF, Pillow, python-pptx, and zipfile to perform document parsing, image processing, and PPTX file generation (SKILL.md, 'Toolchain Policy').
Sanitization: There is no mention of sanitizing, escaping, or validating the text extracted from input documents before it is used to generate slide content and speaker notes.

nature-paper2ppt