Why are curl commands wrapped in bash -c?

Due to a specific environment variable handling behavior in Claude Code, wrapping the curl command in bash -c ensures that your API key is correctly passed when using pipes.

Which OpenAI models are supported by this skill?

The skill supports all current models including GPT-4o, GPT-4o-mini, o1-preview, DALL-E 3, Whisper-1, and various text-to-speech and embedding models.

Does this skill support streaming responses?

Yes, the skill provides examples for enabling stream mode, which returns Server-Sent Events (SSE) for real-time, token-by-token output.

Can I use this skill for image analysis?

Yes, the skill includes specific patterns for Vision tasks using GPT-4o and GPT-4o-mini to analyze image URLs or local files.

How do I authenticate the OpenAI skill?

You must obtain an API key from the OpenAI Platform and set it as an environment variable named OPENAI_API_KEY in your terminal session.

OpenAI API Integration

Name: OpenAI API Integration
Author: vm0-ai

byvm0-ai

•

Data Science & ML

Integrates OpenAI's full suite of models including GPT-4o, DALL-E, and Whisper directly into your terminal via curl.

About

The OpenAI API skill empowers Claude to leverage external AI capabilities from OpenAI through direct curl commands. It provides standardized access to flagship LLMs like GPT-4o and o1, image generation via DALL-E 3, audio transcription with Whisper, and text-to-speech generation. This skill is ideal for developers needing to compare model outputs, generate multimedia assets, create embeddings for RAG systems, or perform vision-based tasks without leaving their CLI environment. It handles complex configurations such as function calling, streaming, and JSON mode while providing clear pricing and rate-limiting guidance.

Key Features

18 GitHub stars
Advanced configurations for JSON mode, function calling, and SSE streaming responses
Pre-configured patterns for generating text embeddings for semantic search and RAG
Support for multimodal tasks including Vision (GPT-4o), DALL-E 3, and Whisper transcription
Standardized curl commands for all major OpenAI endpoints including Chat, Images, and Audio
Built-in model selection guidance for cost-efficiency using GPT-4o-mini and o1 models

Use Cases

Generating and testing vector embeddings for local data analysis and retrieval-augmented generation
Rapid prototyping and benchmarking of AI features across different GPT model generations
Automating content workflows that require image generation, text-to-speech, or audio transcription

About

Key Features

18 GitHub stars
Advanced configurations for JSON mode, function calling, and SSE streaming responses
Pre-configured patterns for generating text embeddings for semantic search and RAG
Support for multimodal tasks including Vision (GPT-4o), DALL-E 3, and Whisper transcription
Standardized curl commands for all major OpenAI endpoints including Chat, Images, and Audio
Built-in model selection guidance for cost-efficiency using GPT-4o-mini and o1 models

Use Cases

Generating and testing vector embeddings for local data analysis and retrieval-augmented generation
Rapid prototyping and benchmarking of AI features across different GPT model generations
Automating content workflows that require image generation, text-to-speech, or audio transcription