tgw-prompt-architect
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill acts as a structural architect for prompt engineering, containing purely instructional content without any functional code, scripts, or external dependencies.
- [DATA_EXFILTRATION]: No network-enabled tools or commands (such as curl, wget, or fetch) are utilized. The instructions explicitly warn against including real customer data, personal information, or private corporate documents in generated outputs.
- [REMOTE_CODE_EXECUTION]: The skill does not perform any environment modifications, package installations, or command executions. All activities are confined to natural language processing and structural evaluation within the agent's context.
- [PROMPT_INJECTION]: No malicious prompt injection patterns or attempts to bypass safety filters were identified. The instructions focus on improving the stability and boundary management of AI prompts.
Audit Metadata