The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash and Task tools to execute commands necessary for its primary function, such as running R scripts for data visualization and performing a 3-pass LaTeX compilation for Beamer slides.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external materials including research papers, slides, and code to extract content for lecture creation.
Ingestion points: Reads from user-provided 'papers and materials' during Phase 0 and Phase 1.
Boundary markers: No explicit delimiter or instruction-guarding markers are defined for the content extracted from external papers.
Capability inventory: The agent has access to Bash, Task, Write, and Edit tools, allowing it to execute scripts and modify the filesystem based on interpreted content.
Sanitization: The workflow lacks automated sanitization for external content, though it includes a human-in-the-loop 'GATE' where the user must approve the structure before drafting slides.

create-lecture