The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands to perform its core tasks, such as executing a local Python script (scripts/extract-pptx.py) for content extraction, installing necessary libraries via pip, and using the open command to display the final presentation to the user. These actions are transparent and necessary for the stated purpose.
[EXTERNAL_DOWNLOADS]: The skill downloads established Python packages (python-pptx, Pillow) from official registries and references well-known typography services (Google Fonts, Fontshare) within the generated HTML code.
[PROMPT_INJECTION]: The skill processes user-supplied data from PowerPoint files and images, creating a potential surface for indirect prompt injection (Category 8). However, the risk is minimal as the data is used strictly for content generation. Mandatory Evidence Chain:
Ingestion points: External content enters the system through the parsing of .pptx files and the agent's visual analysis of provided images.
Boundary markers: The extraction script does not implement explicit delimiters to separate untrusted content from the generation logic.
Capability inventory: The skill has the ability to execute Python scripts, install packages, and write files to the local system.
Sanitization: There is no explicit sanitization or filtering logic applied to extracted text before it is inserted into the final HTML template.

frontend-slides