OpenAI
for LuumenAI

Read stored completions and model details, then run evals as approved steps

Connect OpenAI with an API key and your team gets a working view of the account from the terminal: stored chat completions and their messages, transcriptions, embeddings, moderation checks, file downloads, and model details. Changes — creating evals, starting eval runs, building batches, cancelling work in flight — are previewed and approved before they run.

Book a DemoDownload Luumen

The OpenAI toolbox

126 tools: 66 read, 60 write. Reads answer instantly. Writes require approval by default. Everything is logged.

  • ReadCompact ResponseCompact a conversation or response to reduce token usage.
  • ReadCreate Audio TranscriptionTranscribe audio files to text via OpenAI Audio Transcriptions API.
  • ReadCreate Audio TranslationTranslate audio files to English text via OpenAI Audio Translations API.
  • ReadCreate Chat CompletionCreate a chat completion response from OpenAI models.
  • ReadCreate Completion (Legacy)Generate text completions using OpenAI's legacy Completions API.
  • ReadCreate EmbeddingsGenerate text embeddings via the OpenAI embeddings endpoint.
  • ReadCreate ModerationClassify text and/or image inputs for potentially harmful content via the OpenAI Moderation API.
  • ReadCreate Speech (TTS)Generate text-to-speech audio using OpenAI's Audio API.
  • ReadDownload fileDownload the contents of a specified file by its ID.
  • ReadDownload Video ContentDownload video content (MP4) or preview assets from OpenAI Videos API.
  • ReadGet Chat CompletionRetrieve a stored chat completion.
  • ReadGet Chat Completion MessagesRetrieve messages from a stored chat completion.
  • ReadGet ChatKit threadRetrieve a ChatKit thread by its ID.
  • ReadGet Conversation ItemRetrieve a single item from a conversation.
  • ReadGet EvalRetrieve an evaluation by ID.
  • ReadGet Evaluation RunRetrieve an evaluation run by ID to check status and results.
  • ReadGet Eval Run Output ItemRetrieve a specific output item from an evaluation run by its ID.
  • ReadGet eval run output itemsGet a list of output items for an evaluation run.
  • ReadGet Evaluation RunsGet a paginated list of runs for an evaluation.
  • ReadGet Input Token CountsCalculate input token counts for OpenAI API requests.
  • ReadGet MessageRetrieve a specific message from a thread by its ID.
  • ReadGet ResponseRetrieve a model response by ID.
  • ReadGet Run StepRetrieve a specific run step from an Assistants API run to inspect detailed execution progress, view tool calls, or check message creation.
  • ReadGet Vector StoreRetrieve a vector store by its ID.
  • ReadGet Vector Store FileRetrieve a file from a vector store.
  • ReadGet Vector Store File BatchRetrieve a vector store file batch.
  • ReadGet VideoRetrieve a video generation job by its unique identifier.
  • ReadList AssistantsList assistants to discover the correct assistant_id by name or metadata.
  • ReadList BatchesList your organization's batches.
  • ReadList Chat CompletionsList stored chat completions that were created with the `store` parameter set to true.
  • ReadList ChatKit thread itemsList ChatKit thread items.
  • ReadList container filesList files in a container.
  • ReadList ContainersList containers.
  • ReadList Conversation ItemsList all items for a conversation with the given ID.
  • ReadList enginesList available engines and their basic information.
  • ReadList EvalsList evaluations for a project.
  • ReadList filesRetrieve a list of files uploaded to your organization/project context.
  • ReadList Files in Vector Store BatchList vector store files in a batch.
  • ReadList fine-tunesList your organization's fine-tuning jobs.
  • ReadList fine-tuning job eventsGet status updates for a fine-tuning job.
  • ReadList fine-tuning job checkpointsList checkpoints for a fine-tuning job.
  • ReadList Input ItemsRetrieve input items for a given response from the OpenAI Responses API.
  • ReadList MessagesList messages in an Assistants thread to fetch the assistant's generated outputs after a run completes.
  • ReadList modelsList available models scoped to the current account/organization — some public models may be absent due to permissions.
  • ReadList RunsList runs belonging to a thread.
  • ReadList Run StepsList run steps for an Assistants API run to track detailed execution progress, inspect tool calls, and view message creation events.
  • ReadList SkillsList skills.
  • ReadList ChatKit ThreadsList ChatKit threads with pagination and filtering.
  • ReadList Vector Store FilesList files in a vector store.
  • ReadList Vector StoresList vector stores to discover available vector stores by name or metadata.
  • ReadList VideosList all video generation jobs.
  • ReadRetrieve assistantRetrieve details of a specific assistant.
  • ReadRetrieve BatchRetrieve a batch by ID to check its status, progress, and results.
  • ReadRetrieve containerRetrieve details of a specific container by its ID.
  • ReadRetrieve container fileRetrieve metadata for a specific file in a container.
  • ReadRetrieve container file contentRetrieve the content of a file within a container.
  • ReadRetrieve engineRetrieve details of a specific engine.
  • ReadRetrieve fileRetrieve information about a specific file.
  • ReadRetrieve fine-tuning jobRetrieve information about a fine-tuning job.
  • ReadRetrieve modelRetrieve details of a specific model, confirming its metadata (ownership, created date) and verifying access under your org — a model appearing in OPENAI_LIST_MODELS does not guarantee access.
  • ReadRetrieve runRetrieve an Assistants run by ID to check status, errors, and usage.
  • ReadRetrieve threadRetrieve metadata of a specific thread by its ID — does not include message bodies or assistant replies (those require a completed run and separate message listing).
  • ReadRetrieve Vector Store File ContentRetrieve the parsed contents of a vector store file.
  • ReadRun graderRun a grader to evaluate model performance on a given sample.
  • ReadSearch Vector StoreSearch a vector store for relevant chunks based on a query and file attributes filter.
  • ReadValidate grader configurationValidate a grader configuration for fine-tuning jobs.
  • WriteAdd Upload PartAdd a part (chunk of bytes) to an Upload object. Approval by default
  • WriteCancel batchCancel an in-progress batch. Approval by default
  • WriteCancel evaluation runCancel an ongoing evaluation run. Approval by default
  • WriteCancel ResponseCancel a background model response by its ID. Approval by default
  • WriteCancel RunCancel a run that is currently in progress. Approval by default
  • WriteCancel uploadCancel an upload. Approval by default
  • WriteCreate BatchCreate and execute a batch from an uploaded file of requests. Approval by default
  • WriteCreate ContainerCreate a new container with configurable memory, expiration, file access, and network policies. Approval by default
  • WriteCreate Container FileCreate a file in a container. Approval by default
  • WriteCreate ConversationCreate a new conversation for multi-turn interactions. Approval by default
  • WriteCreate Conversation ItemsCreate items in a conversation with the given ID. Approval by default
  • WriteCreate EvalCreate an evaluation structure for testing a model's performance. Approval by default
  • WriteCreate Evaluation RunCreate a new evaluation run for testing model configurations. Approval by default
  • WriteCreate fine-tuning jobCreate a fine-tuning job which begins the process of creating a new model from a given dataset. Approval by default
  • WriteGenerate ImageGenerate an image via the OpenAI Images API and return hosted image asset URL and metadata. Approval by default
  • WriteEdit ImageCreate edited or extended images via OpenAI Images Edit API. Approval by default
  • WriteCreate Image VariationCreate a variation of a given image using the OpenAI Images API. Approval by default
  • WriteCreate MessageCreate a new message in a specific thread. Approval by default
  • WriteCreate Realtime CallCreate a Realtime API call over WebRTC and receive the SDP answer needed to complete the peer connection. Approval by default
  • WriteCreate Realtime Client SecretCreate an ephemeral client secret for authenticating Realtime API connections. Approval by default
  • WriteCreate Realtime SessionCreate an ephemeral API token for client-side Realtime API applications. Approval by default
  • WriteCreate Realtime Transcription SessionCreate an ephemeral API token for realtime transcriptions via the Realtime API. Approval by default
  • WriteCreate ResponseGenerate a one-shot model response via the Responses API. Approval by default
  • WriteCreate RunCreate a run on a thread with an assistant. Approval by default
  • WriteCreate SkillCreate a skill from uploaded files. Approval by default
  • WriteCreate ThreadCreate a new thread. Approval by default
  • WriteCreate Thread And RunCreate a thread and run it in one request. Approval by default
  • WriteCreate UploadCreate an intermediate Upload object for large file uploads. Approval by default
  • WriteCreate Vector StoreCreate a new vector store. Approval by default
  • WriteCreate Vector Store FileCreate a vector store file by attaching a File to a vector store. Approval by default
  • WriteCreate vector store file batchCreate a vector store file batch. Approval by default
  • WriteCreate VideoCreate a video using Sora models via the OpenAI Videos API. Approval by default
  • WriteCreate Video RemixCreate a video remix from an existing generated video using OpenAI's Video API. Approval by default
  • WriteDelete assistantDelete a specific assistant by its ID. Approval by default
  • WriteDelete chat completionDelete a stored chat completion by its ID. Approval by default
  • WriteDelete containerDelete a specific container by its ID. Approval by default
  • WriteDelete container fileDelete a file from a container. Approval by default
  • WriteDelete conversationDelete a conversation by its ID. Approval by default
  • WriteDelete conversation itemDelete an item from a conversation with the given IDs. Approval by default
  • WriteDelete evaluationDelete a specific evaluation by its ID. Approval by default
  • WriteDelete evaluation runDelete an evaluation run. Approval by default
  • WriteDelete fileDelete a file by its ID after confirming the target. Approval by default
  • WriteDelete messageDelete a message from a thread. Approval by default
  • WriteDelete responseDelete a model response with the given ID. Approval by default
  • WriteDelete skillDelete a specific skill by its ID. Approval by default
  • WriteDelete threadDelete a thread by its ID. Approval by default
  • WriteDelete Vector StoreDelete a vector store. Approval by default
  • WriteDelete Vector Store FileDelete a vector store file. Approval by default
  • WriteDelete videoDelete a video by its ID. Approval by default
  • WriteModify AssistantModify an existing assistant. Approval by default
  • WriteModify MessageModify an existing message's metadata in a thread. Approval by default
  • WriteModify RunModify a run's metadata. Approval by default
  • WriteModify threadModify an existing thread's metadata. Approval by default
  • WriteModify Vector StoreModify an existing vector store. Approval by default
  • WriteSubmit Tool Outputs to RunSubmit tool call outputs to continue a run that requires action. Approval by default
  • WriteUpdate Chat CompletionUpdate metadata for a stored chat completion. Approval by default
  • WriteUpdate ConversationUpdate a conversation's metadata. Approval by default
  • WriteUpdate EvalUpdate certain properties of an evaluation (name and metadata). Approval by default
  • WriteUpdate Vector Store File AttributesUpdate custom attributes on a vector store file. Approval by default
  • WriteUpload fileUpload a file for use across OpenAI endpoints. Approval by default

One prompt, start to finish

What a governed OpenAI run looks like inside Luumen.

Questions

How does LuumenAI connect to OpenAI?

Authorize once with API token. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.

Can LuumenAI change things in OpenAI on its own?

Read actions answer immediately. Anything that writes — add upload part, cancel batch, cancel evaluation run, cancel response, and more — is shown as a plan and requires approval by default, including the 21 actions classified as destructive. Administrators configure that per tool, so you decide exactly which actions can ever run unattended.

Who gets access to the integration?

You decide. Actions are granted per agent, skill, and team, and per environment — production is not staging. Read access can be broad while writes stay narrow.

Is there an audit trail?

Every call to OpenAI — read or write, approved or declined — is recorded with the actor, the input, and the result, and can be linked to the ticket or change record.

All integrations
Adobe Marketing Agent MCPAdobe Experience Platform context, sandbox switches, and async task statusAffindaRead parsed documents, collections, and annotations — change them with approvalAgent MailGive the agent its own mailbox for reading threads and sending updatesAI/ML APIInspect assistants, threads, and runs — and change them as approved steps

Put OpenAI to work with Luumen

Connect in minutes. Every action scoped, approved, and audited from day one.

Book a DemoDownload Luumen