context-injection
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is purely instructional, providing guidelines for prompt engineering. It includes security best practices, such as using structured delimiters to prevent prompt injection and identifying insecure code patterns (e.g., SQL injection) in provided examples to demonstrate the effectiveness of grounded context.- [NO_CODE]: No executable code, scripts, or external tool configurations are present in the provided file. The content is restricted to documentation and examples in Markdown format.
Audit Metadata