Automation Architect
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- [NO_CODE] (SAFE): The skill is composed entirely of YAML metadata and Markdown documentation. It lacks any functional code (Python, JavaScript, Shell) or executable binaries.- [PROMPT_INJECTION] (SAFE): No patterns for instruction override, system prompt extraction, or safety filter bypass were detected. The instructions are focused on guiding the agent through a design process.- [DATA_EXFILTRATION] (SAFE): The skill does not contain logic to read sensitive files (e.g., credentials, SSH keys) or initiate network connections to external domains.- [EXTERNAL_DOWNLOADS] (SAFE): There are no references to external package managers (pip, npm) or remote script execution (curl | bash).- [OBFUSCATION] (SAFE): No hidden text, encoded strings (Base64), or homoglyphs were found in the metadata or body content.
Audit Metadata