investor-due-diligence
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and process information from external, third-party data sources such as Crunchbase, LinkedIn, and Chinese business news platforms (36Kr, ITJuzi).
- Ingestion points: Data is gathered from external web searches and news reports at runtime.
- Boundary markers: The skill defines a strict output structure but lacks explicit instructions to ignore embedded prompts within the external text.
- Capability inventory: No dangerous system capabilities (shell execution, file system writes) are present in the skill definition.
- Sanitization: No explicit sanitization of external text is defined.
- [SAFE]: No obfuscation, persistence mechanisms, or unauthorized privilege escalation commands were detected. The skill uses standard research patterns and clearly defines its evaluation logic.
Audit Metadata