The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and parses content from external LaTeX preamble files (.sty, .tex) and configuration files (.latexmkrc) provided in a project directory. There is a potential for these files to contain embedded instructions designed to influence the agent's behavior during analysis or when generating the compliance report. While the skill uses structured parsing into semantic blocks, it lacks explicit boundary markers or sanitization logic to mitigate adversarial content within the processed files.
Ingestion points: paper/your-template.sty, paper/your-bib-template.sty, paper/settings.tex, main.tex, and .latexmkrc files located in the user-provided project path.
Boundary markers: None explicitly defined to separate project content from agent instructions.
Capability inventory: Access to Edit, Write, and restricted Bash commands (ls, latexmk, lualatex, biber, cmp, cp, head).
Sanitization: The instructions focus on semantic extraction but do not specify filtering or escaping of the raw text extracted from project files.

latex-template