hunt-oauth
Fail
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs users to use the command
curl [URL] | python3 -m json.toolto format JSON data retrieved from a remote server. Whilejson.toolis a built-in Python module for JSON processing and not for code execution, the pattern of piping unvalidated network data directly into a language interpreter is an unsafe practice that can lead to remote code execution if the command is mistyped or if the remote source provides malicious content. - [EXTERNAL_DOWNLOADS]: The skill provides various payloads and test cases that involve external URLs (e.g.,
https://evil.com,https://legit.com/logout?next=https://evil.com). These are intended for security researchers to test for open redirects, path traversals, and OAuth token leakage. Although flagged by scanners as malicious or phishing-related, they are contextually appropriate for the skill's purpose as a vulnerability hunting tool.
Recommendations
- HIGH: Downloads and executes remote code from: https://target.com/.well-known/openid-configuration - DO NOT USE without thorough review
- Contains 5 malicious URL(s) - DO NOT USE
Audit Metadata