ln-23-system-design-proposal-builder
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional prompts and checklists for system architecture design. It does not contain any executable scripts, shell commands, or network operations.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository content (manifests, architecture artifacts). Ingestion points: repository search and stakeholder input. Boundary markers: the execution contract requires explicit verification status (PROVEN/CLEARED/UNPROVEN) for each design claim. Capability inventory: the skill is document-focused and lacks code execution or network exfiltration tools. Sanitization: none. This surface is considered safe due to the skill's restricted operational context and rigorous verification logic.
Audit Metadata