flow-domain-creator
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for architectural documentation and does not exhibit any malicious behavior or security risks.
- [PROMPT_INJECTION]: The skill processes untrusted input from
vision.mdto generate downstream documentation. - Ingestion points: Reads
vision.mdat the start of the workflow (file:SKILL.md). - Boundary markers: Absent; the content is analyzed directly without explicit delimiters or warnings to ignore embedded instructions.
- Capability inventory: File writing to the local
domains/directory (file:SKILL.md). - Sanitization: No sanitization of ingested content before inclusion in the final output template.
Audit Metadata