bellingcat-image-video
Pass
Audited by Gen Agent Trust Hub on Jul 12, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No prompt injection attempts, safety bypass instructions, or adversarial framing patterns were detected. The instructions are purely functional and align with the stated purpose of identifying OSINT tools.
- [DATA_EXFILTRATION]: No evidence of hardcoded credentials, sensitive file path access, or unauthorized network operations. The listed URLs point to legitimate investigative tools and documentation.
- [EXTERNAL_DOWNLOADS]: The skill does not perform any automated downloads or package installations. It contains Markdown links to external websites for manual use or reference by the agent.
- [REMOTE_CODE_EXECUTION]: There are no patterns involving the execution of remote scripts, piped shell commands, or dynamic code evaluation.
- [COMMAND_EXECUTION]: The skill contains no shell commands or subprocess calls. It is an informational Markdown file.
- [CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or private secrets were found within the skill file.
- [SAFE]: The skill serves as a reference guide for legitimate OSINT tools. References to well-known services (Google, Amazon, Microsoft) and established investigative organizations (Bellingcat, RAND Corporation) are documented neutrally and follow safe practices.
Audit Metadata