venue-templates
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute several local Python scripts, including generate_schematic.py, query_template.py, customize_template.py, and validate_format.py. It also invokes standard LaTeX compilation tools like pdflatex and latexmk to generate and verify document compliance.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data, creating a potential surface for indirect instruction execution.
- Ingestion points: The skill reads and validates user-provided manuscript files (.tex and .pdf) and processes natural language descriptions to generate scientific diagrams.
- Boundary markers: There are no explicit delimiters or instructions provided to the agent to help distinguish between data and potentially malicious embedded instructions in the processed files.
- Capability inventory: The skill is scoped with Bash, Read, Write, and Edit tools, allowing it to execute scripts and modify the file system.
- Sanitization: The instructions do not define any sanitization or validation routines for user-supplied strings before they are passed as arguments to the local Python scripts via the shell.
Audit Metadata