What problem does it solve? Adding a new AI provider monitor to the OpenRouter codebase requires touching multiple files with consistent patterns — schema definitions, StandardMonitor instantiation, auth headers, and index registration. This Skill provides a step-by-step checklist with ready-to-use templates so nothing is missed. ## Core Features & Use Cases - Three config templates: Covers simple public endpoints, API-key-authenticated providers, and providers with non-standard response shapes. - Schema guidance: Shows how to extend StandardModelDataSchema with provider-specific fields and normalize values like hf_slug, max_completion_tokens, and supported_sampling_parameters. - Registration checklist: Walks through updating the all-monitors index, adding API key env vars, and optionally creating vendor tests. - Use Case: When onboarding a new inference provider like NextBit or IO.net, follow this Skill to create the monitor config, wire up authentication, and register it in getProviderMonitors(). ## Quick Start Add a provider monitor for the provider named X with model listing API at its /v1/models endpoint, using the appropriate template for its auth and response shape.
