private-company-research
Pass
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill retrieves information from established professional and financial platforms such as SEC filings, LinkedIn, and SimilarWeb for research purposes.
- [PROMPT_INJECTION]: The skill processes data from external websites, which presents a surface for indirect prompt injection.
- Ingestion points: Public web content retrieved during the data collection phase.
- Boundary markers: Not explicitly defined in the sub-agent prompt instructions.
- Capability inventory: Access to sub-agent creation and file writing tools.
- Sanitization: No specific filtering of external content is mentioned.
- [COMMAND_EXECUTION]: The skill uses local shell commands and project-specific tools to execute its research workflow and manage output files.
Audit Metadata