architectural-decision-records
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious behavior or high-risk patterns were detected in the skill's instructions or templates.
- [EXTERNAL_DOWNLOADS]: The skill references template documentation from public GitHub repositories and GitHub Pages, such as the MADR and Nygard templates. These references are for informational purposes and do not involve downloading executable scripts or packages.
- [INDIRECT_PROMPT_INJECTION]: The skill acts as a text-processing engine that ingests user input to generate documentation. While it handles external data, the risks are inherent to its primary purpose and limited to local file generation.
- Ingestion points: User-provided architectural context and decision details in templates (e.g., templates/madr-full.md).
- Boundary markers: Absent.
- Capability inventory: File system access for updating preferences.md and generating new markdown-based ADRs.
- Sanitization: None detected.
Audit Metadata