payloads-headless
Fail
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: CRITICALDATA_EXFILTRATIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill provides functional JavaScript payloads designed to read local files like /etc/passwd and exfiltrate them via HTTP POST requests to external domains. It also includes the domain 'evil.net' in its documentation, which automated scanners have flagged as a phishing threat.
- [EXTERNAL_DOWNLOADS]: The documentation points to raw scripts hosted on GitHub Gists, including 'ripWCMN.py' and 'chrome_remote_debug_lfi.py'. These are third-party, unverified resources.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of arbitrary external web content, presenting a significant attack surface for indirect prompt injection. Ingestion points: Web content rendered by a headless browser (SKILL.md). Boundary markers: None provided. Capability inventory: File system access, network communication, and remote browser control. Sanitization: No input sanitization or safety checks are included in the workflow.
- [SAFE]: The skill is explicitly labeled for authorized security research and penetration testing purposes and contains a legal notice.
Recommendations
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata