The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/generate_infographic.py acts as a wrapper that executes a secondary Python script (generate_infographic_ai.py) using subprocess.run(). While this follows the standard pattern for modularizing logic, it involves the execution of local shell commands.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to openrouter.ai and perplexity.ai to interface with language and image generation models. These are well-known technology services required for the skill's primary function of infographic generation and research.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. In scripts/generate_infographic_ai.py, the _enhance_prompt_with_research method interpolates external data fetched via the Perplexity API directly into the generation prompt.
Ingestion points: External data enters the context from the research_topic function in scripts/generate_infographic_ai.py via the Perplexity Sonar Pro model.
Boundary markers: The skill uses simple text headers such as "RESEARCHED DATA AND FACTS" but lacks robust escaping or "ignore previous instructions" delimiters to isolate untrusted data.
Capability inventory: The skill has capabilities to write files to the local system (images and logs), execute subprocesses (via the wrapper script), and perform network operations.
Sanitization: There is no evidence of sanitization or validation of the content returned from the external research API before it is passed to the next stage of the AI pipeline.

infographics