regex-vs-llm-structured-text
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security threats identified. The skill consists of documentation and Python code snippets for building text-parsing pipelines.
- [SAFE]: The Python implementation uses standard libraries such as
reanddataclasses. It does not perform any sensitive system operations, network requests to unknown domains, or file system modifications. - [SAFE]: The LLM integration in the
validate_with_llmfunction is used for data validation of extracted text. While it processes untrusted input, the skill is a framework intended for developer use and does not possess capabilities (such as shell access or credential access) that would make indirect prompt injection a high-risk concern.
Audit Metadata