# Openai > OpenAI API — generate text, images, speech, transcriptions, and embeddings using GPT, DALL-E, Whisper, and TTS models. Supports chat completions, vision, function calling, and file management via `robomotion openai`. Do NOT use for Claude, Gemini, or direct LLM conversation. - Skill: `robomotionio/openai` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add robomotionio/openai` - Raw SKILL.md: https://api.skillmd.com/api/skills/robomotionio/openai/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: AI & ML - License: Apache-2.0 - Author: robomotionio (https://skillmd.com/u/robomotionio) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/robomotionio/openai --- # OpenAI The `robomotion openai` CLI calls the OpenAI API for AI operations. It generates text with GPT models (including vision and function calling), creates images with DALL-E/GPT-Image, synthesizes and transcribes audio, generates embeddings, moderates content, and manages files. ## When to use - Generate text or chat completions with GPT models (including tool use) - Create or edit images with DALL-E and GPT-Image models - Convert text to speech or transcribe/translate audio with Whisper - Generate embeddings, moderate content, and manage uploaded files ## Prerequisites - `robomotion` CLI installed - Package installed: `robomotion install openai` - OpenAI API key configured via Robomotion vault ## IMPORTANT: Session Mode Required for Multi-Step Operations Each CLI command runs as a **separate process** — connection IDs from `connect` do NOT persist across calls. You MUST use `--session` on `connect` and pass `--session-id` to all subsequent commands. ## Workflow 1. Install (once): `robomotion install openai` 2. Connect with session: ``` robomotion openai connect_openai --session --output json # → {"outConnectionId":"","session_id":""} ``` 3. Use the returned `connection-id` and `session-id` in all subsequent commands: ``` robomotion openai generate_text --connection-id "" --user-prompt --session-id "" --output json ``` 4. Disconnect when done: ``` robomotion openai disconnect_openai --connection-id "" --session-id "" --output json ``` **Always** append `--output json` to get structured JSON results. ## Commands Reference - `robomotion openai connect_openai --session --output json` Connect to OpenAI API and create a client session for AI operations - `robomotion openai disconnect_openai --connection-id --session-id "" --output json` Disconnect from OpenAI API and release the client session - `robomotion openai generate_text --connection-id --you-are-a-helpful-assistant- --user-prompt --history --tool-message --image-url --images [--model] [--custom-model] [--number-of-generations] [--temperature] [--top-p] [--max-tokens] [--frequency-penalty] [--presence-penalty] [--stop-sequences] [--reasoning-effort] [--image-detail] [--functions] [--tools-choice] [--seed] [--user] [--120] --session-id "" --output json` Generate text completions using OpenAI language models with optional vision and tool calling - `robomotion openai generate_image --connection-id --prompt [--model] [--number-of-images] [--size] [--quality] [--style] [--background] [--output-format] [--user] [--120] --session-id "" --output json` Generate images using OpenAI DALL-E and GPT-Image models - `robomotion openai edit_image --connection-id --image-path --mask-path --prompt [--model] [--number-of-images] [--size] [--user] [--120] --session-id "" --output json` Edit an existing image based on a text prompt using DALL-E or GPT-Image models - `robomotion openai create_image_variation --connection-id --image-path [--model] [--number-of-images] [--size] [--user] [--120] --session-id "" --output json` Create variations of an existing image using DALL-E 2 - `robomotion openai generate_speech --connection-id --text [--model] [--voice] [--format] [--1-0] [--instructions] [--120] --session-id "" --output json` Convert text to speech using OpenAI TTS models - `robomotion openai transcribe_audio --connection-id --audio-file [--model] [--language] [--prompt] [--0] [--120] --session-id "" --output json` Transcribe audio to text using OpenAI Whisper and GPT-4o transcription models - `robomotion openai translate_audio --connection-id --audio-file [--model] [--prompt] [--0] [--120] --session-id "" --output json` Translate audio to English text using OpenAI Whisper model - `robomotion openai generate_embedding --connection-id --text [--model] [--custom-model] [--dimensions] [--user] --session-id "" --output json` Generate vector embeddings from text using OpenAI embedding models - `robomotion openai generate_batch_embeddings --connection-id --texts [--model] [--custom-model] [--dimensions] [--user] --session-id "" --output json` Generate vector embeddings for multiple texts in a single batch request - `robomotion openai similarity --connection-id --search-embeddings-csv --content-embeddings-csv [--5] --session-id "" --output json` Calculate cosine similarity between embeddings to find most relevant content - `robomotion openai moderate --connection-id --text [--model] --session-id "" --output json` Check text content for policy violations using OpenAI moderation models - `robomotion openai list_models --connection-id --session-id "" --output json` List available OpenAI models with their details - `robomotion openai retrieve_model --connection-id --gpt-5-mini --session-id "" --output json` Retrieve detailed information about a specific OpenAI model - `robomotion openai upload_file --connection-id --file-path [--purpose] --session-id "" --output json` Upload a file to OpenAI file storage for use with assistants or fine-tuning - `robomotion openai list_files --connection-id [--purpose-filter] --session-id "" --output json` List all files uploaded to OpenAI file storage - `robomotion openai retrieve_file_metadata --connection-id --file-id --session-id "" --output json` Retrieve metadata about a file in OpenAI file storage - `robomotion openai download_file --connection-id --file-id --session-id "" --output json` Download a file from OpenAI file storage to local filesystem - `robomotion openai delete_file --connection-id --file-id --session-id "" --output json` Delete a file from OpenAI file storage ## Environment - ROBOMOTION_API_TOKEN (if vault credentials are needed)