OpenAI
Put OpenAI to work on your own records, inside your rules, with a person approving what it changes.
27 ways to read, 11 ways to write, each write switched on by you.
Quality checks
Built and tested with example responses.
- Ready to connect: doneEvery listed action is built in.
- Built and tested with example responses: doneEach action is built from the provider’s published API and tested against example responses, errors, rate limits and paging.
Connected, approved, proven.
The things teams do most with OpenAI, an example flow with a person in the loop, and what Nexra keeps as proof.
Top actions
- Create batchApproval by rule
- Create embeddingsApproval by rule
- List audit logsRead
Example flow
- OpenAI · List audit logsDone
- OpenAI · Create batchNeeds approval
- Nexra · Record the decision and result in the evidence chainProven
Example workflow. Routine runs go through on their own. Only a change your rules flag, such as one over a limit you set, waits for the right person, who sees the exact change and the rule before it reaches OpenAI.
What gets logged
- Whether it ran on its own or who approved it, and under which rule
- A fingerprint of the exact request sent to OpenAI
- OpenAI's reply: what it created or changed
- One entry in the evidence chain, linked to the one before
Try this in the sandbox
Run the whole flow with test data in your sandbox, then connect your own account when you are ready to go live.
For developers
Technical detail
Embeddings, extraction, moderation, files, batches, fine-tuning, vector stores, stored responses and organization usage, cost and audit reporting from OpenAI.
OpenAI API docs- Source actions
- 27
- Destination actions
- 11
- Data types
- 22
What you can map
Reads are always available. Writes are off until you switch them on for a connection, and governed writes follow your policies.
Audit logs1
- List audit logsSourceIncremental
Organization audit events (API key, project, user, role and login changes), filterable by time, project and event type. Requires the Admin API key and audit logging enabled.
Batches3
- Create batchDestinationNot retried automaticallyApproval before production
Starts a 24-hour batch from an uploaded JSONL file (purpose batch) against one endpoint. Batch requests are billed at the batch rate.
- Get batchSource
Status, request counts, output file and usage for one batch.
- List batchesSource
Batch jobs in the project, newest first, with request counts and file IDs.
Chat completions2
- Get stored chat completionSource
One stored chat completion with its first choice and usage.
- List stored chat completionsSource
Chat completions created with store=true, optionally for one model.
Costs1
- List organization costsSource
Daily spend in the billing currency, one record per bucket result, optionally grouped by project or line item. Requires the Admin API key.
Embeddings1
- Create embeddingsDestinationNot retried automaticallyApproval before production
Vector embeddings for one text or a list of texts, one record per input in order.
Extraction1
- Extract structured dataDestinationNot retried automaticallyApproval before production
Extracts a JSON object that conforms to your schema from input text using the Responses API with Structured Outputs.
Files2
- Get file metadataSource
Metadata for one file (no contents).
- List filesSource
File metadata in the project (no file contents).
Fine tuning checkpoints1
- List fine-tuning checkpointsSource
Checkpoints saved during one fine-tuning job, with step metrics.
Fine tuning events1
- List fine-tuning job eventsSource
Status and metrics events for one fine-tuning job, newest first.
Fine tuning jobs2
- Get fine-tuning jobSource
Status, hyperparameters, trained tokens and result files for one job.
- List fine-tuning jobsSource
Fine-tuning jobs in the project with status, base model and resulting model.
Models2
- Get modelSource
One model, including fine-tuned models, with owner and shutdown date.
- List ModelsSource
List operation for Models.
Moderation1
- Classify content for moderationDestinationNot retried automaticallyApproval before production
Classifies text with the moderation model and returns flagged categories and scores.
Project API keys2
- Get project API keySource
Metadata for one project API key. Requires the Admin API key.
- List project API keysSource
API key metadata (redacted value, owner, last used) for a project. Requires the Admin API key.
Project rate limits1
- List project rate limitsSource
Per-model rate limits configured for a project. Requires the Admin API key.
Project service accounts1
- List project service accountsSource
Service accounts in a project with their role. Requires the Admin API key.
Projects2
- Get projectSource
One project. Requires the Admin API key.
- List projectsSource
Projects in the organization. Requires the Admin API key.
Responses1
- Get responseSource
A stored or background model response with status, output text and usage (for example to collect a background response).
Text1
- Create TextDestinationNot retried automaticallyApproval before production
Create operation for Text.
Usage1
- List organization usageSource
Usage by time bucket for one API surface (completions, embeddings, images, vector stores and more), one record per bucket result. Requires the Admin API key.
Vector store file batches2
- Add files to vector store in a batchDestinationNot retried automaticallyApproval before production
Attaches up to 500 uploaded files to a vector store in one batch (billable storage).
- Get vector store file batchSource
Progress of a vector store file batch.
Vector store files4
- Add file to vector storeDestinationNot retried automaticallyApproval before production
Attaches an uploaded file to a vector store; OpenAI chunks and embeds it (billable storage).
- Get vector store fileSource
Ingestion status, usage and attributes for one file in a vector store.
- List vector store filesSource
Files attached to a vector store with ingestion status and attributes.
- Update vector store file attributesDestinationProvider duplicate protectionApproval before production
Replaces the filterable attributes of a file in a vector store.
Vector stores5
- Create vector storeDestinationNot retried automaticallyApproval before production
Creates a vector store, optionally with already-uploaded files. Storage above the free allowance is billed per GB per day.
- Get vector storeSource
One vector store with status, file counts and expiry.
- List vector storesSource
Vector stores in the project with file counts and storage usage.
- Search vector storeDestinationNot retried automaticallyApproval before production
Semantic search over a vector store's chunks with optional attribute filters; one record per matching chunk group.
- Update vector storeDestinationProvider duplicate protectionApproval before production
Renames a vector store or changes its expiry policy or metadata.
Access and limits
- Authentication
- Project API key (plus optional Admin API key)
- Plans and access
- OpenAI API platform projects with billing enabled. Model availability and rate limits follow the project's usage tier and the key's project permissions. Usage, cost, project, API key and audit log reads need an organization Admin API key created by an Organization Owner.
- Test environment
- There is no sandbox; use a separate project with a low spend limit for testing.
- Rate limits
- Requests and tokens per minute per model, by usage tier; batch queue limits per model. HTTP 429 with x-ratelimit-* and retry-after headers.
- Pagination
- Lists use after cursors (last_id) while has_more is true; usage and cost reports use next_page tokens over time buckets.
- Provider events
- OpenAI webhooks for background responses, batches, fine-tuning jobs and evals. Flows run on demand, on a schedule, or when a scheduled check finds new or changed records.
Before you build a flow
- Authenticated model discovery and bounded text generation. Select a model available to your key. Generation sends mapped content to the provider and incurs provider charges. Automatic retries are disabled; inspect usage before repeating an uncertain request. No agents, tool calls, image generation or embeddings.
- Embeddings, extraction, batch creation, vector store creation, vector store file attachment and vector store search are billable and are never retried automatically; check usage before repeating an uncertain request.
- Extraction sends store=false so OpenAI does not retain the response for later retrieval.
- File upload (multipart), the Uploads API and file content downloads are not implemented: the recipe engine sends JSON or form bodies only and returns JSON. Batch input files must be uploaded outside Nexra; batch results are read by file ID.
- Fine-tuning jobs, events and checkpoints are read only; creating, pausing, resuming and cancelling jobs are not implemented.
- The Assistants API and its threads, messages and runs are deprecated by OpenAI and are not implemented; use the Responses API and vector stores instead.
- Organization usage, costs, projects, project API keys, service accounts, rate limits and audit logs require the optional Admin API key. API keys are read only (redacted values); creating, rotating and deleting keys, user and role changes and spend limit changes are intentionally excluded.
- Usage and cost reports accept one project, API key and group-by value per request.
- Deleting models, files, vector stores, vector store files, responses and chat completions is intentionally excluded.
- Moderation is free but sends content to OpenAI; it is a destination step that must be enabled per connection.
Developer reference
- Create the OpenAI credentials described above with the narrowest permissions your flows need.
- In Nexra, open Connect, then Connections, then New connection, choose OpenAI, enter its settings and credentials, and tick only the write actions you need. Connection guide
- Test the connection, then use its actions as flow sources and destinations. Flow guide
Machine-readable detail: GET /api/public/connect/connectors/openai and the Connections API.