hunt-race-condition
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill includes Bash and Python code snippets intended for security testing, such as using
curlin parallel loops oraiohttpfor concurrent requests. These are strictly for the primary purpose of authorized bug hunting and utilize placeholders for target URLs and session cookies. - [EXTERNAL_DOWNLOADS]: References authoritative security research from PortSwigger and Flatt Security. These links are provided for documentation and technical reference and do not involve the automated download or execution of untrusted scripts.
- [SAFE]: Implements a 'Mandatory confirmation gate' that requires the agent to verify the user's target scope and written authorization before providing any active commands. This enforces a read-only posture until explicit permission is confirmed.
Audit Metadata