csp-bypass-advanced
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a technical knowledge base for security testing. It documents known CSP vulnerabilities and bypass vectors without including any executable malicious payloads.
- [PROMPT_INJECTION]: The 'AI LOAD INSTRUCTION' provides legitimate context for the agent regarding the skill's technical scope and does not attempt to override system instructions or safety filters.
- [DATA_EXFILTRATION]: While the skill describes data exfiltration techniques (e.g., DNS prefetch, WebRTC), these are provided as educational examples for security researchers. No actual exfiltration logic or hardcoded credentials were found.
- [REMOTE_CODE_EXECUTION]: References to external CDNs like jsdelivr or unpkg are used to illustrate how trusted origins can be abused in CSP bypasses; the skill itself does not download or execute remote code.
- [DYNAMIC_CONTEXT_INJECTION]: The skill does not use any shell execution placeholders or dynamic context patterns.
Audit Metadata