blockchain-red-team
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary purpose is offensive security research, which is clearly stated in its metadata and instructions. All content aligns with this purpose.
- [SAFE]: No hardcoded credentials, sensitive file access, or network exfiltration patterns were found.
- [SAFE]: There are no automated remote code execution patterns or unverifiable dependency installations.
- [SAFE]: No obfuscation techniques (Base64, zero-width characters, etc.) were detected.
- [SAFE]: While the skill instructs the agent to research external sources such as audit reports and CTF writeups, which technically creates a surface for potential indirect prompt injection, this risk is inherent to the research task and the skill does not provide automated ingestion scripts. Evidence: Ingestion points are external reports; Boundary markers are absent; Capability inventory includes PoC generation; Sanitization is not specified.
Audit Metadata