The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface as it ingests untrusted textbook data to influence generated narratives and image prompts. * Ingestion points: reads docs/course-description.md, mkdocs.yml, and docs/learning-graph/learning-graph.csv. * Boundary markers: No explicit delimiters or instructions are used to isolate textbook data from the skill's operational instructions. * Capability inventory: Filesystem writes (docs/stories/), system command execution (sips, sed), and network requests to the Google Gemini API. * Sanitization: No input validation or filtering of textbook content is performed before processing.
[COMMAND_EXECUTION]: The Python scripts generate-images.py and verify-images.py utilize subprocess.check_output to execute the system utility sips for image dimension verification.
[EXTERNAL_DOWNLOADS]: The skill requires the google-genai Python library and makes network requests to official Google Gemini API endpoints for image generation.
[DATA_EXFILTRATION]: Local textbook metadata and narrative prompts are transmitted to external Google Gemini API endpoints as part of the expected generation workflow.

story-generator