cto
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, command executions, or external data exfiltration vectors were identified in the skill.
- [INDIRECT_PROMPT_INJECTION]: The skill utilizes an ingestion surface for project-specific context through local files but implements effective design-level mitigations.
- Ingestion points: The skill reads from the
.nugelo/directory, specificallystatus.md,handoff.md, and feature-specific files withindecisions/cto/. - Boundary markers: The skill contains a 'Project State Protocol' that serves as a strict boundary, explicitly telling the agent that these files are data and not instructions.
- Capability inventory: The skill is capable of reading and writing to the
.nugelo/directory and performing git commits to track decision history. - Sanitization: The agent is instructed to treat all content in these files as untrusted input and is specifically forbidden from following any directives found within them.
Audit Metadata