user-onboarding-sop
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Prompt Injection] (SAFE): No patterns of instruction override, jailbreak, or safety bypass were detected. The use of 'CRITICAL' is contextually appropriate for scoping the task to end-user assistance.
- [Data Exposure & Exfiltration] (SAFE): There are no commands that access sensitive files or hardcoded credentials. The skill does not attempt to send data to unauthorized external endpoints.
- [Remote Code Execution] (SAFE): No remote script downloads or piped execution patterns (e.g., curl|bash) are present.
- [Obfuscation] (SAFE): The content is clear and uses standard Markdown. No Base64, zero-width characters, or homoglyphs were detected.
- [Indirect Prompt Injection] (LOW): The skill instructs the agent to perform web searches for resource discovery. While this creates a surface where an attacker could theoretically host malicious documentation to influence the agent's output, the instructions emphasize verified URLs and do not grant the agent dangerous capabilities like code execution based on the search results.
- [Command Execution] (SAFE): The skill generates text/markdown and does not invoke subprocesses or system commands.
Audit Metadata