methods-section-writer
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
- [SAFE]: The skill is a specialized text-generation tool for academic writing. It contains no executable code, shell scripts, or capability configurations that would allow for system modification or network access.
- [SAFE]: All external resources and references directed to well-known academic standards (e.g., EQUATOR Network, CONSORT, STROBE) and the author's official GitHub repository, which is a legitimate vendor resource.
- [SAFE]: The core workflow incorporates mandatory safety constraints that prevent the fabrication of ethics approval identifiers, statistical results, or p-values, utilizing a placeholder system to ensure integrity.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied protocols and analysis workflows. While this ingestion of external data constitutes a theoretical attack surface, the risk is mitigated by the skill's lack of action-oriented tools (e.g., no tools for file writing or network requests are defined in the frontmatter).
- [METADATA_POISONING]: The skill includes a static self-evaluation report (
eval_report_methods-section-writer_result.json) containing self-referential claims about its own safety and performance. These claims were evaluated as data points rather than authoritative conclusions, and no deceptive metadata was found.
Audit Metadata