workflow-greenfield-init
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious code, external downloads, or data exfiltration patterns were detected in the skill instructions or metadata.
- [PROMPT_INJECTION]: The skill uses authoritative directives (e.g., "MUST", "Do NOT skip") to ensure the agent follows the specific sequence of the project initialization workflow. These are natural operational instructions for a complex task and do not represent a safety bypass.
- [PROMPT_INJECTION]: The skill ingests user input (the prompt) as the primary context for the
greenfield-initworkflow. While this serves as an ingestion point for potential indirect prompt injection, it is the fundamental purpose of the skill and relies on the agent's internal guardrails for security. - Ingestion points: User prompt (SKILL.md)
- Boundary markers: Absent
- Capability inventory: Triggers a chain of internal commands including
/web-research,/deep-research,/scaffold, and/test(SKILL.md) - Sanitization: None specified in the skill body
Audit Metadata