regex-vs-llm-structured-text
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The
validate_with_llmfunction inSKILL.mdis susceptible to indirect prompt injection. * Ingestion points: Untrusted content via theraw_textparameter is passed directly into a prompt template. * Boundary markers: The prompt templatef"Text: {raw_text}\nDraft: {current_data}"lacks explicit delimiters like XML tags or backticks to isolate the untrusted data from the instructions. * Capability inventory: No high-risk capabilities like file system writes or shell command execution are defined in the provided snippets. * Sanitization: No sanitization or filtering is applied to the input text before it is sent to the LLM.
Audit Metadata