The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/generate_infographic.py uses subprocess.run to call a secondary generation script. It uses an argument list for execution, which is a secure practice that avoids shell injection vulnerabilities.- [EXTERNAL_DOWNLOADS]: The skill connects to openrouter.ai to interact with Perplexity and Gemini APIs for data research and image generation. These requests target a well-known and expected service provider for such tasks.- [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface in scripts/generate_infographic_ai.py where external research data is used to build prompts. \n
Ingestion points: Research results from Perplexity Sonar are incorporated into the image generation prompt in the generate_iterative method. \n
Boundary markers: Research content is concatenated with the prompt instructions without delimiters or markers to distinguish it from agent commands. \n
Capability inventory: The skill has the capability to generate images and write them to the filesystem via the OpenRouter API. \n
Sanitization: There is no filtering or validation of the text retrieved from the research phase before it is used in the prompt instructions.

infographics