What problem does it solve? Developers building with OpenAI APIs often rely on outdated or unofficial information when choosing models, configuring Codex, or migrating prompts. This Skill routes questions to authoritative OpenAI documentation sources so answers stay current and citable. ## Core Features & Use Cases - Official docs lookup: Searches and fetches pages from the developers.openai.com MCP server, with fallback to official-domain web search when MCP tools are unavailable. - Codex self-knowledge: Fetches and caches the Codex manual via a Node helper script, generating an outline for targeted section reads on setup, skills, plugins, MCP, hooks, and automations. - Model selection and migration: Resolves the latest model guidance, preserves explicitly requested model targets, and applies bundled upgrade and prompting guides as fallbacks. - Use Case: When asked to migrate an integration to the latest GPT model, the Skill resolves the current model info, fetches the migration and prompting guides, and applies narrow model-string and prompt updates while flagging out-of-scope changes as blocked. ## Quick Start Ask how to configure a Codex feature or which OpenAI model to use for your application and request citations from the official docs.
