openai▌

Transcribe an audio file via OpenAI’s /v1/audio/transcriptions endpoint. Set OPENAI_BASE_URL to use an OpenAI-compatible proxy or local gateway.

Access current OpenAI developer documentation with MCP tools and model-selection guidance. \n \n Prioritizes the OpenAI Developer Docs MCP server for all queries; falls back to web search only on official OpenAI domains if MCP returns no results \n Includes reference files for model selection, GPT-5.4 upgrade planning, and prompting guidance; always verifies recommendations against live docs before answering \n Covers seven core OpenAI products: Apps SDK, Responses API, Chat Completions, Codex,

Local speech-to-text transcription using OpenAI's Whisper CLI without API keys. \n \n Transcribes audio files (MP3, M4A, and other formats) directly on your machine with no external API calls required \n Supports multiple model sizes (tiny, base, small, medium, large, turbo) with automatic caching to ~/.cache/whisper on first run \n Offers transcription and translation tasks with configurable output formats (TXT, SRT, JSON, VTT) \n Requires only the whisper CLI binary, installable via Homebrew o

Provide authoritative, current guidance from OpenAI developer docs using the developers.openai.com MCP server. Always prioritize the developer docs MCP tools over web.run for OpenAI-related questions. Only if the MCP server is installed and returns no meaningful results should you fall back to web search.

This skill covers using OpenAI's embedding API with GrepAI for high-quality, cloud-based embeddings.

Build ChatGPT apps with MCP servers on Cloudflare Workers, extending ChatGPT with custom tools and interactive HTML/JS widgets. \n \n Implements JSON-RPC 2.0 MCP protocol with tool registration, execution, and widget resource serving via ASSETS binding \n Requires CORS allowlist for https://chatgpt.com , ui://widget/ URI prefix for widgets, and text/html+skybridge MIME type \n Prevents 14 documented issues including CORS blocks, widget 404s, MIME type misconfigurations, SSE timeouts, and Next.js

Build stateful chatbots with OpenAI Assistants API v2, including Code Interpreter, File Search (10k files), and Function Calling. \n \n Three core tools: Code Interpreter for Python execution and file processing, File Search for semantic RAG with vector stores (10,000 files max), and Function Calling for custom tool integration \n Manages four main objects—Assistants (configured AI with instructions), Threads (persistent conversation containers), Messages (with file attachments), and Runs (async

Stateful agentic conversations with preserved reasoning, server-side tools, and automatic state management. \n \n Preserves model reasoning across turns (+5% performance on TAUBench), eliminating manual history tracking and improving multi-turn interactions \n Built-in server-side tools: Code Interpreter, File Search, Web Search, DALL-E, and MCP integration without backend round trips \n Automatic conversation state management with 90-day expiration; 40-80% better cache utilization than Chat Com

Build text and voice agents with tools, multi-agent handoffs, guardrails, and human-in-the-loop patterns. \n \n Supports text agents, realtime voice agents with WebRTC, and multi-agent workflows with automatic delegation via handoffs \n Define tools using Zod schemas for type-safe parameter validation; includes structured output support for predictable JSON responses \n Implements input/output guardrails for safety validation, human approval workflows for sensitive operations, and streaming supp

Complete reference for OpenAI's stateless APIs including Chat Completions, embeddings, images, audio, and moderation. \n \n Supports GPT-5 series (with reasoning_effort control), GPT-4o multimodal, o3 reasoning models, and legacy GPT-4 variants; streaming, function calling, and structured JSON outputs included \n Embeddings API for RAG with custom dimensions (256–3072), DALL-E 3 image generation, Whisper transcription, and TTS with 11 voices \n Batch API for 50% cost savings on large-scale proce