The Agent Skills Directory

[SAFE]: No malicious code, obfuscation, or unauthorized data exfiltration patterns were detected in the skill instructions or associated scripts. The script uses official libraries and follows standard API usage patterns.\n- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes external markdown files to generate visual content.\n
Ingestion points: The script scripts/generate_slide_images.py reads the full content of all .md files located in the directory specified by the --prompts-dir argument.\n
Boundary markers: Absent. The raw content from the markdown files is directly interpolated into the generation prompt sent to the Gemini API.\n
Capability inventory: The script can read files from the local filesystem, write PNG files to an output directory, and communicate with the Google Gemini API.\n
Sanitization: Absent. User-provided content is not filtered or sanitized before being included in the model prompt.

slide-renderer