arch-patterns
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill does not contain any executable scripts, remote downloads, or obfuscated content. It functions primarily as a set of instructions for an AI agent to provide software design advice.- [SAFE]: Authentication is handled via a standard environment variable ($OPENCLAW_API_KEY), which is a secure practice for CLI-based tools and does not expose credentials.- [SAFE]: While the skill processes user-supplied code snippets for analysis and refactoring, it lacks the capabilities, such as file system write access or arbitrary command execution, that would make indirect prompt injection a significant threat.
Audit Metadata