OpenWarrant — AI Text Detection

You'd reach for this when you need to analyze text for AI generation fingerprints. It exposes operations that return probability scores indicating whether prose was likely written by an LLM, along with specific tells that triggered the detection. Useful for content moderation workflows, academic integrity checks, or any pipeline where you need to programmatically flag synthetic text. The server runs on Stipple's infrastructure and connects via streamable HTTP, so you can call it from Claude without managing your own detection models or keeping up with evolving LLM signatures.