The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documents routine shell commands for compiling LaTeX documents, specifically pdflatex, bibtex, and lualatex for generating PDF presentations from templates.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of research data and content into structured slides, which establishes a vulnerability surface where untrusted data could influence agent output.
Ingestion points: Research text, findings, and figures provided by the user for presentation development as described in Stage 1 and Stage 3 of the workflow.
Boundary markers: The skill does not provide explicit delimiters or instructions to the agent to ignore potentially malicious embedded instructions in the research data.
Capability inventory: Shell command execution for LaTeX compilation is recommended in the protocol guidelines.
Sanitization: There are no specific validation or sanitization routines implemented for external user-provided data.

scientific-slides