The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted markdown content, creating a potential vector for indirect prompt injection attacks where malicious instructions could be embedded in the data being converted.
Ingestion points: Untrusted data enters the skill via the markdown_content parameter in scripts/markdown_processor.py and scripts/epub_generator.py.
Boundary markers: There are no explicit boundary markers or instructions to ignore embedded commands within the processed markdown content.
Capability inventory: The skill has the capability to write to the file system using the ebooklib library to save the resulting EPUB file, as seen in scripts/epub_generator.py.
Sanitization: The skill implements HTML escaping for content and metadata using the _escape_html and escape_html methods to prevent common injection issues in the generated XHTML structure.

markdown-to-epub-converter