The Agent Skills Directory

[SAFE]: The skill's behavior is entirely consistent with its stated purpose. No malicious patterns such as obfuscation, credential theft, persistence, or data exfiltration were identified during the analysis.
[COMMAND_EXECUTION]: The skill uses several shell commands including pdflatex, bibtex, biber, latexmk, grep, file, and iconv. These tools are standard for LaTeX document management and are used locally to automate compilation, search build logs, and handle file encoding conversions within the project scope.
[PROMPT_INJECTION]: The skill processes untrusted document content and compiler logs, representing an indirect prompt injection surface.
Ingestion points: LaTeX source files (.tex), bibliography files (.bib), and compilation logs (build.log).
Boundary markers: None; file content is processed as raw text without explicit delimiters.
Capability inventory: Execution of local LaTeX compilation engines and text-processing utilities; write access to project .tex files.
Sanitization: The skill includes specific guardrails to prevent modification of critical document sections (e.g., metadata, bib keys, and comments) which mitigates the risk of unintended modifications caused by input data.

latex-rescue