osint
Fail
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: CRITICALEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides numerous commands to interact with external sources. Specifically, the file
guides/dark-web-intelligence-guide.mdcontains a reference tohttp://juhanurmihxlp77nkq76byazcldy2hlmovfu2epvl5ankdibsot4csyd.onion(Ahmia search engine), which has been flagged by automated scanners as a malicious phishing site. - [DATA_EXFILTRATION]: The skill includes instructions and payloads (e.g., in
payloads.mdandguides/dark-web-intelligence-guide.md) for interacting withpastebin.com,paste.ee, andpsbdmp.ws. Automated scanners recognize these platforms as common destinations for data exfiltration. The skill's focus on monitoring these sites for sensitive information creates a potential risk for data leakage if used without oversight. - [INDIRECT_PROMPT_INJECTION]: The skill processes extensive data from untrusted sources, creating an indirect prompt injection attack surface.
- Ingestion points: Data is collected from search engines, social media (LinkedIn, Twitter), dark web forums, and public code repositories (GitHub, Gitee) through various OSINT tools.
- Boundary markers: The instructions do not define clear delimiters or provide warnings for the agent to treat external content as untrusted data rather than potential instructions.
- Capability inventory: The skill provides access to the
Bashtool and includes numerous Python and Bash scripts for network and file operations. - Sanitization: There is no evidence of sanitization or filtering for the external content retrieved by the tools before it is processed by the agent.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute powerful system utilities (e.g.,
masscan,nmap,subfinder,spiderfoot) via theBashtool and provides templates for automated reconnaissance pipelines that execute sequences of shell commands. - [DYNAMIC_EXECUTION]: Guides such as
guides/automated-osint-pipeline-guide.mdandguides/2026-03-21-information-gathering-tools.mdprovide code that utilizessubprocess.runto execute system commands andurllib.request.urlopento interact with remote services, posing a risk of dynamic code execution of untrusted payloads.
Recommendations
- AI detected serious security threats
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata