interface-designer
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions establish a 'Company Framework' focusing on technical integrity and security protocols. It mandates the use of specific MCP (Model Context Protocol) loops for sequential thinking and context verification, which are standard practices for controlled agent behavior.
- [SAFE]: The security protocols explicitly address 'Multi-Agent Hygiene' by enforcing least privilege and directing the agent to treat external data (such as web content) as hostile. It mandates the redaction of PII and secrets before sharing context with subagents, demonstrating a security-first design.
- [SAFE]: Terminal operations are managed through a defined 'ExecutionProxy' (rtk prefix), and persistent data is stored in localized documentation paths (Obsidian
docs/departments/). These patterns represent standard, organized project management rather than unauthorized persistence or exfiltration mechanisms.
Audit Metadata