proxyman-traffic-debugging
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides comprehensive instructions for using Proxyman MCP tools to inspect and manage local network traffic. The instructions emphasize privacy by requiring the agent to respect the application's built-in redaction settings for sensitive data like tokens and passwords.
- [COMMAND_EXECUTION]: The skill documents capabilities for process injection (Terminal and Electron) and system proxy modifications. These operations are presented as standard debugging features of the Proxyman suite and are coupled with instructions for the agent to seek explicit user consent before execution.
- [DATA_EXPOSURE]: While the skill processes network traffic which may contain sensitive information, it mitigates exposure risk by instructing the agent to use Proxyman's redaction safeguards and requiring user-verified paths for any flow exports to disk.
- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it processes untrusted data from network flows. However, the risk is addressed through the application's redaction mechanisms and the skill's instructions to treat the MCP server's structured schemas as the source of truth.
Audit Metadata