crewai
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates for creating agents that ingest untrusted data from external sources and user-provided inputs, creating a surface for indirect prompt injection attacks.
- Ingestion points: The
research_taskandwriting_taskinSKILL.mdutilize a{topic}variable and fetch data using external tools. - Boundary markers: The instruction templates do not demonstrate the use of delimiters or specific warnings to ignore instructions embedded within the processed data.
- Capability inventory: The skill configures agents with capabilities to perform network searches using
SerperDevToolandWebsiteSearchTool. - Sanitization: No sanitization, filtering, or validation logic is present in the provided examples to protect against malicious payloads in retrieved content.
Audit Metadata