zscaler-zpa
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, Base64 encoding, or hidden command execution were detected in the skill instructions.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill lists sensitive credentials (
ZSCALER_CLIENT_ID,ZSCALER_CLIENT_SECRET,ZSCALER_CUSTOMER_ID) as prerequisites to be supplied by the user via environment variables. This demonstrates a secure approach to secret management by avoiding hardcoded values. - [INDIRECT_PROMPT_INJECTION]: The skill includes tools that retrieve data from the Zscaler API (e.g.,
get_application_segment,list_access_policies). While this creates a potential surface for indirect prompt injection from the API's response, the skill itself does not contain any exploitable logic, and this is a standard risk for API-integrated tools. - [COMMAND_EXECUTION]: The skill references an external server (
zscaler-mcp), which is consistent with the Model Context Protocol (MCP) for extending agent capabilities. No unauthorized shell execution or dangerous piped commands were found.
Audit Metadata