reviewing-architecture-doc
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is instructional and does not contain functional code, shell commands, or network requests. It defines a manual workflow for an agent to evaluate the quality of text-based documents against a specific 10-condition bar.
- [PROMPT_INJECTION]: The skill processes untrusted input in the form of architecture documents and ADR files, which constitutes an indirect prompt injection surface. A malicious document could attempt to influence the agent's output. However, the impact is minimal as the skill does not possess sensitive capabilities like filesystem modification or network exfiltration. Ingestion points: External architecture documents and linked ADR files (SKILL.md, Step 1). Boundary markers: None explicitly defined. Capability inventory: Limited to text analysis and emission of a formatted verdict. Sanitization: None described in the instructions.
Audit Metadata