Integration

OpenAI integration

AI models API for text generation, embeddings, image generation, video generation, audio processing, realtime audio, and more. Generate text and structured outputs with the Responses API, create embeddings for semantic search, generate and edit images with GPT Image models, generate videos with Sora, transcribe audio with Whisper and GPT-4o models, convert text to speech, create realtime audio sessions, and leverage built-in tools like web search, code interpreter, and file search. Perfect for building AI-powered applications, chatbots, content generation tools, and intelligent automation systems.

Connect OpenAI How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in OpenAI.Built from the 78 OpenAI tools below, on the connection you add once.

Common OpenAI use cases

resource management
cleanup
storage management
debugging
document Q&A
file management
storage cleanup
stopping incorrect batches
verification
error recovery

What you can do with OpenAI

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Vectorstore, 6 operations

create_vector_storeCreate vector storePOSTCreates a vector store for semantic search with file_search tool.
delete_vector_storeDelete vector storeDELETEDeletes a vector store.
get_vector_storeGet vector storeGETRetrieves details of a specific vector store.
list_vector_storesList vector storesGETLists all vector stores in your organization.
search_vector_storeSearch vector storePOSTSearches a vector store for relevant chunks based on a query.
update_vector_storeUpdate vector storePOSTUpdates a vector store's name, expiration policy, or metadata.

Assistant, 5 operations

create_assistantCreate assistantPOSTCreates a new AI assistant with custom instructions, tools, and capabilities. Defines persistent assistant configurations that can be reused across multiple conversation threads.
delete_assistantDelete assistantDELETEPermanently deletes an assistant.
get_assistantGet assistantGETRetrieves details of a specific assistant by ID.
list_assistantsList assistantsGETRetrieves a paginated list of all assistants created in your OpenAI account. Returns assistant configurations including names, instructions, models, and enabled tools for management purposes.
update_assistantUpdate assistantPOSTUpdates an existing assistant's configuration.

File, 5 operations

delete_fileDelete fileDELETEDeletes an uploaded file from your account.
get_fileGet fileGETRetrieves information about a specific file by ID.
get_file_contentGet file contentGETReturns the contents of the specified file for download.
list_filesList filesGETLists all files uploaded to your OpenAI account.
upload_fileUpload filePOSTUploads a file for assistants, fine-tuning, or batch processing.

Message, 5 operations

create_messageCreate messagePOSTCreates a new message in a conversation thread.
delete_messageDelete messageDELETEDeletes a message from a thread.
get_messageGet messageGETRetrieves a specific message from a thread.
list_messagesList messagesGETLists all messages in a conversation thread.
update_messageUpdate messagePOSTUpdates message metadata.

Response, 5 operations

cancel_responseCancel responsePOSTCancels an in-progress model response.
create_responseCreate responsePOSTCreates a model response with optional built-in tools (web search, file search, code interpreter, computer use). The primary API for new OpenAI integrations, replacing Chat Completions and Assistants for advanced use cases.
delete_responseDelete responseDELETEDeletes a stored model response by ID.
get_responseGet responseGETRetrieves a model response by ID with full output, usage stats, and metadata.
list_response_input_itemsList response input itemsGETLists input items for a response including messages, images, and files provided as input.

Run, 5 operations

cancel_runCancel runPOSTCancels a run that is in progress.
get_runGet runGETRetrieves details of a specific run.
list_runsList runsGETLists all runs for a conversation thread.
run_assistantRun assistantPOSTRuns an assistant on a thread to generate a response.
submit_tool_outputsSubmit tool outputsPOSTSubmits tool outputs for a run requiring action.

Video, 5 operations

create_videoCreate videoPOSTCreates a video generation job from a text prompt using Sora models. Supports size, duration, and input reference images for first-frame control.
delete_videoDelete videoDELETEPermanently deletes a video from OpenAI storage.
download_videoDownload videoGETDownloads completed video as MP4, thumbnail as WebP, or spritesheet as JPG.
get_videoGet videoGETRetrieves video generation job status including progress percentage, completion state, and any errors.
list_videosList videosGETLists all video generation jobs with pagination and sorting options.

Batch, 4 operations

cancel_batchCancel batchPOSTCancels an in-progress batch job.
create_batchCreate batchPOSTCreates a batch job for processing multiple requests asynchronously with 50% cost discount.
get_batchGet batchGETRetrieves details of a specific batch job.
list_batchesList batchesGETLists all batch jobs in your organization.

Finetuningjob, 4 operations

cancel_fine_tuning_jobCancel fine tuning jobPOSTCancels a fine-tuning job that is in progress.
create_fine_tuning_jobCreate fine tuning jobPOSTCreates a fine-tuning job to customize a model on your data.
get_fine_tuning_jobGet fine tuning jobGETRetrieves details of a specific fine-tuning job.
list_fine_tuning_jobsList fine tuning jobsGETLists all fine-tuning jobs in your account.

Thread, 4 operations

create_threadCreate threadPOSTCreates a new conversation thread for multi-turn interactions with assistants. Optionally initializes with messages to provide conversation context or history.
delete_threadDelete threadDELETEDeletes a conversation thread and all its messages.
get_threadGet threadGETRetrieves a conversation thread by ID.
update_threadUpdate threadPOSTUpdates thread metadata.

Vectorstorefile, 4 operations

create_vector_store_fileCreate vector store filePOSTAdds a file to a vector store for indexing.
delete_vector_store_fileDelete vector store fileDELETERemoves a file from a vector store.
get_vector_store_fileGet vector store fileGETRetrieves a file from a vector store.
list_vector_store_filesList vector store filesGETLists files in a vector store.

Model, 3 operations

delete_modelDelete modelDELETEDeletes a fine-tuned model from your organization.
get_modelGet modelGETRetrieves detailed information about a specific OpenAI model by its identifier. Returns ownership, permissions, and metadata useful for understanding model capabilities and access rights.
list_modelsList modelsGETRetrieves a list of all AI models available to your OpenAI account. Returns model identifiers, ownership information, and capabilities for discovering which models can be used with different API endpoints.

Upload, 3 operations

cancel_uploadCancel uploadPOSTCancels an in-progress Upload.
complete_uploadComplete uploadPOSTCompletes the Upload and creates a usable File object.
create_uploadCreate uploadPOSTCreates an Upload object for uploading large files in parts (up to 8GB).

Vectorstorefilebatch, 3 operations

cancel_vector_store_file_batchCancel vector store file batchPOSTCancels an in-progress vector store file batch.
create_vector_store_file_batchCreate vector store file batchPOSTCreates a batch to add up to 500 files to a vector store.
get_vector_store_file_batchGet vector store file batchGETRetrieves details of a vector store file batch.

Image, 2 operations

create_imageCreate imagePOSTGenerates original images from text descriptions using GPT Image models. Transforms natural language prompts into visual artwork with control over size, quality, and output format for creative and commercial applications.
edit_imageEdit imagePOSTModifies existing images based on text instructions using GPT Image models. Allows selective editing of image regions defined by a mask with control over background, quality, and output format.

Chat completion, 1 operation

create_chat_completionCreate chat completionPOSTGenerates conversational AI responses using GPT models with multi-turn dialogue support. Accepts a sequence of messages with roles (system, user, assistant) and returns contextually appropriate completions with fine-tuned control over creativity a...

Completion, 1 operation

create_completionCreate completionPOSTGenerates text completions from a prompt using legacy completion models. Provides direct text continuation without conversational structure, useful for simple text generation tasks with customizable sampling parameters.

Embedding, 1 operation

create_embeddingCreate embeddingPOSTConverts text into numerical vector representations (embeddings) for semantic similarity, search, and clustering applications. Returns dense floating-point arrays that capture meaning and context of input text.

Finetuningcheckpoint, 1 operation

list_fine_tuning_checkpointsList fine tuning checkpointsGETLists checkpoints created during fine-tuning.

Finetuningevent, 1 operation

list_fine_tuning_eventsList fine tuning eventsGETLists events for a fine-tuning job.

Moderation, 1 operation

create_moderationCreate moderationPOSTAnalyzes text and image content for policy violations across multiple categories including hate speech, violence, and self-harm. Returns detailed category scores for content filtering and safety applications.

Realtime session, 1 operation

create_realtime_sessionCreate realtime sessionPOSTCreates a Realtime API session for live voice-to-voice conversations with AI. Returns a client secret for WebSocket or WebRTC connection.

Runstep, 1 operation

get_run_stepGet run stepGETRetrieves details of a specific run step.

Speech, 1 operation

create_speechCreate speechPOSTConverts text to natural-sounding speech audio using OpenAI's TTS models. Offers multiple voice options, controllable speech with gpt-4o-mini-tts, and various output formats for voiceovers and accessibility.

Step, 1 operation

list_run_stepsList run stepsGETLists the steps taken during a run.

Threadrun, 1 operation

create_thread_and_runCreate thread and runPOSTCreates a thread and runs an assistant in a single request.

Transcription, 1 operation

create_transcriptionCreate transcriptionPOSTConverts audio files to text using Whisper or GPT-4o transcription models. Supports multiple languages, speaker diarization, and word-level timestamps for accurate speech-to-text conversion.

Translation, 1 operation

create_translationCreate translationPOSTTranslates audio in any supported language directly to English text using Whisper. Performs speech recognition and translation in a single step for efficient multilingual audio processing.

Uploadpart, 1 operation

add_upload_partAdd upload partPOSTAdds a part (chunk up to 64MB) to an Upload.

Vectorstorebatchfile, 1 operation

list_vector_store_batch_filesList vector store batch filesGETLists files in a vector store file batch.

Use OpenAI from Claude Code, Cursor or Codex

Connect OpenAI in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its OpenAI operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect OpenAI

Create your workspace
Sign up free. Nothing to install.
Connect OpenAI once
Add your OpenAI connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.

OpenAI integration FAQ

What can you do with the OpenAI integration?

LiveContext exposes 78 operations on OpenAI, covering vectorstore, assistant, file and message. Common uses include resource management, cleanup and storage management.