retro
Pass
Audited by Gen Agent Trust Hub on Jul 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional text and does not include any executable code, scripts, or remote dependencies.
- [SAFE]: No network requests, shell command execution, or access to sensitive local files or credentials were identified.
- [SAFE]: Indirect prompt injection surface analysis: The skill defines a process for reading untrusted data (user complaints, transcripts) and generating documentation. This is a low-risk analytical task with no exploitable capabilities such as code execution or network exfiltration. Ingestion points: final artifacts, QA evidence, and user complaints listed in SKILL.md; Boundary markers: none provided; Capability inventory: writing and refreshing local documentation in SKILL.md; Sanitization: none specified.
Audit Metadata