regex-vs-llm-structured-text

Pass

Audited by Gen Agent Trust Hub on Feb 24, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The validate_with_llm function in SKILL.md is susceptible to indirect prompt injection.
  • Ingestion points: External content passed via the original_text variable is directly embedded into the prompt string.
  • Boundary markers: The implementation lacks distinct delimiters (like XML tags) or specific instructions to the model to ignore potential commands within the input text.
  • Capability inventory: While the skill focuses on extraction, compromised extraction results could negatively impact any application logic relying on this data.
  • Sanitization: There is no evidence of input validation or sanitization before the data is processed by the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 24, 2026, 06:53 PM