What problem does it solve? Adding a new AI model provider to LobeHub touches many disconnected layers—runtime adapters, model banks, environment configuration, icons, translations, and bilingual docs—and missing any one of them produces broken or misleading integrations. This Skill provides a complete, evidence-driven checklist so the provider works end to end for users. ## Core Features & Use Cases - Full Integration Contract: Guides registration of provider IDs, provider cards, model cards, runtime factories, RouterRuntime adapters, and authentication plumbing (API key or OAuth). - Branding and Ordering: Covers icon matching in @lobehub/icons, deliberate placement in DEFAULT_MODEL_PROVIDER_LIST, and localized provider descriptions in English and Chinese. - Documentation Workflow: Produces bilingual usage guides, environment variable references, and Docker configuration derived from implemented behavior. - Validation Harness: Uses providerTestUtils and focused package tests to verify streaming, structured generation, error localization, and settings UI behavior. - Use Case: When asked to add a new provider like a local inference server or a Responses-only API, follow the integration map and local-providers references to implement, brand, document, and validate the provider without inventing unsupported capabilities. ## Quick Start Ask the AI to add a new model provider to LobeHub by name, for example requesting it to integrate a specific provider with its models, icons, and documentation.
