g5
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user-provided text for analysis. While it uses the
[Text]: {text}delimiter as a boundary marker, it lacks explicit sanitization for instructions embedded within the text. However, as the skill's capabilities are limited to providing analysis and reporting (passive phase) and do not include high-risk tools like network requests or file writes, this surface represents a standard operational risk for this use case. - [DYNAMIC_EXECUTION]: The skill includes a Python code snippet for LaTeX syntax validation using the
latex2ommlpackage. This code is used for static analysis of text and does not execute user-provided code or perform dangerous runtime compilation. - [SAFE]: Metadata and instructions are consistent with the stated purpose of academic auditing. No obfuscation, persistence mechanisms, or privilege escalation attempts were detected. References to project-specific files like
.research/decision-log.yamlare part of the Model Context Protocol (MCP) fallback mechanism and do not involve sensitive system data.
Audit Metadata