bmel:containsAny¶
Category: LLM & AI Observability · Returns: bmel:boolean
bmel:containsAny(text: bmel:string, patterns: bmel:array)
Description¶
Returns true if the text contains at least one string from the provided array. Case-insensitive by default. Useful for detecting hallucination markers, toxicity patterns, PII keywords, or known failure signatures in LLM outputs without requiring a full embedding-based classifier.
Arguments¶
| Parameter | Type | Required | Description |
|---|---|---|---|
text | bmel:string | ✅ | The text to search within. |
patterns | bmel:array | ✅ | Array of strings to search for. Returns true on the first match found. |
Example¶
bmel:containsAny({chat:Response Payload}.$.content, ["I cannot", "I'm unable", "As an AI"])