crewai
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION] (LOW): The skill templates demonstrate patterns vulnerable to indirect prompt injection. 1. Ingestion points: The {topic} variable is interpolated directly into agent goals and task descriptions in config/agents.yaml and config/tasks.yaml. 2. Boundary markers: No delimiters or 'ignore' instructions are used to separate user-provided data from agent instructions. 3. Capability inventory: The skill references SerperDevTool and WebsiteSearchTool, which allow the agent to fetch external content. 4. Sanitization: No sanitization or validation for the {topic} input is illustrated.
- [SAFE] (SAFE): No critical or high-risk patterns such as hardcoded credentials, malicious network requests, or obfuscated code were found in the skill content.
Audit Metadata