architect
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill adheres to secure design principles and does not contain malicious instructions, obfuscation, or unauthorized access patterns.- [DATA_EXFILTRATION]: While the skill involves reading project files to understand existing architecture (grounding), it does not access sensitive system files (e.g., credentials, SSH keys) or perform any network exfiltration.- [COMMAND_EXECUTION]: The skill coordinates high-level agent capabilities (explore, implement, verify) within the pstack framework but does not execute arbitrary or dangerous shell commands.- [PROMPT_INJECTION]: No attempts to subvert AI safety guidelines or bypass ethical filters were detected. The use of model roles (e.g., critic, judgment) is a legitimate architectural pattern for design synthesis.
Audit Metadata