The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use the Bash tool to execute a suite of local Python scripts, including generate_slide_image.py for slide creation, slides_to_pdf.py for assembly, and validate_presentation.py for quality checks. These are standard functional components of the skill's workflow.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, potentially untrusted data by attaching user-provided figures and charts (e.g., from figures/ or results/ directories) to AI generation prompts using the --attach flag. This creates a surface where instructions embedded in image metadata or text within images could influence the model's output.
Ingestion points: Visual data and research figures read from the local workspace (e.g., figures/*.png, results/*.jpg).
Boundary markers: The documentation does not specify explicit boundary markers or "ignore" instructions to separate attached visual content from the generation prompt.
Capability inventory: The skill has access to Bash for script execution, Read for data ingestion, and Write for generating presentation files.
Sanitization: No explicit sanitization or filtering of the content within the attached images is mentioned.
[EXTERNAL_DOWNLOADS]: The skill facilitates communication with the OpenRouter API (openrouter.ai), a well-known service, to perform the core AI-driven slide generation tasks.

scientific-slides