What problem does it solve? Evaluating whether an AI inference provider trains on your prompts, retains your data, or silently routes requests to third-party hosts requires digging through dense privacy policies and terms of service. This Skill structures that investigation into a repeatable audit covering provider archetype, free-tier privacy discrimination, zero data retention availability, and jurisdiction exposure. ## Core Features & Use Cases - Provider Classification: Identifies whether a platform is a first-party model builder, dedicated inference host, or intermediary gateway forwarding payloads to third parties. - Privacy Matrix Generation: Produces a standardized comparison table covering training-on-payloads defaults, paywalled privacy controls, retention periods, ZDR availability, and data residency. - Opt-Out Action Plans: Delivers step-by-step hardening instructions covering account settings, API parameters, cookie controls, and local self-hosted alternatives. - Use Case: Before adopting a free LLM API tier, run an audit to discover that free-tier prompts are used for training by default, then follow the generated opt-out steps or switch to local GGUF weights. ## Quick Start Audit the privacy and data retention policies of the OpenRouter inference provider and give me the full opt-out action plan.
