sci-paper
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of instructional markdown files and LaTeX templates for academic publishing. Analysis did not detect any evidence of prompt injection, data exfiltration, obfuscation, or persistence mechanisms.
- [COMMAND_EXECUTION]: The skill provides standard documentation for compiling LaTeX manuscripts using
pdflatexandbibtexin thereferences/latex_template.mdfile. These commands are informational and appropriate for the skill's stated purpose of producing PDF documents. - [DYNAMIC_EXECUTION]: The skill workflow involves the generation and compilation of LaTeX source code into PDF artifacts. This runtime compilation is a standard utility of the
pdflatextool and is used here in a low-risk, domain-appropriate context. - [INDIRECT_PROMPT_INJECTION]: The skill defines a system for processing and formatting scientific research content provided by users, which constitutes an ingestion surface for indirect prompt injection.
- Ingestion points: User-provided research data and drafting instructions processed via the workflow defined in SKILL.md.
- Boundary markers: None explicitly defined in the templates or instructions.
- Capability inventory: The skill facilitates local file writing and the use of standard LaTeX compilation tools (
pdflatex,bibtex) as noted in references/latex_template.md. - Sanitization: No validation or sanitization of user-provided content is implemented in the instructional templates.
Audit Metadata