wazza-check
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes report data from the Wazza service. It includes a specific safeguard instructing the agent to treat report text exclusively as data and never as instructions, which effectively mitigates indirect prompt injection risks from the processed files.
- [CREDENTIALS_UNSAFE]: Authentication is handled via an environment variable ($WZ_KEY). The skill includes mandatory instructions forbidding the agent from printing, writing, or requesting this key, adhering to secure credential management practices.
- [EXTERNAL_DOWNLOADS]: The skill instructions guide the user to manually add the Wazza MCP server using a vendor-controlled URL (api-pro.wazza.ai). This is a standard setup procedure for extending agent capabilities with trusted vendor tools.
Audit Metadata