insights-deck
Pass
Audited by Gen Agent Trust Hub on May 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Potential for Indirect Prompt Injection.
- Ingestion points: Processes HTML files located in
/tmp/orlog/insights/(SKILL.md). - Boundary markers: Absent. The skill lacks explicit delimiters or instructions for the agent to ignore embedded commands within the processed HTML.
- Capability inventory: Uses
Bash(latexmk, cp, mkdir),Read,Write, andEdit(SKILL.md). - Sanitization: Absent. The skill does not describe any sanitization of the data extracted from the HTML reports before using it to generate LaTeX source.
- [COMMAND_EXECUTION]: Executes shell commands (
ls,mkdir,cp,latexmk) to manage files and compile documents. - [COMMAND_EXECUTION]: Generates a
.latexmkrcconfiguration file that includes a Perlsystem()call, which is executed during the LaTeX build process.
Audit Metadata