Use OpenAI from Claude
Add the OpenAI connection to a Space that runs Claude, then use it from every channel in that Space.
Connect OpenAI once, then control which Spaces can use its approved tools and data without pasting credentials into prompts or threads.
Add the OpenAI connection to a Space that runs Claude, then use it from every channel in that Space.
Add the OpenAI connection to a Space that runs Codex, then use it from every channel in that Space.
Connect your own OpenClaw to Sidekick while type.com provides the conversations, permissions, connections, skills, and automations.
OpenAI's platform covers model inference along with assistants, files, batches, evals, and fine-tuning. Adding an OpenAI connection to type.com lets Claude or Codex operate that account for your engineers: uploading big files in parts, stopping batches, evaluation runs, background responses, or assistant runs that should not keep going, and reporting what changed in a shared thread.
Connect OpenAI once, then decide which Spaces can use it in threads, skills, automations, and coding work.
When a batch was submitted with the wrong prompt or model, cancel it quickly from the thread before it burns more of your budget.
Cancel OpenAI batch batch_7Hk2 and confirm its status after cancellation.
Cancel an evaluation run that started against the wrong dataset, then capture in the thread what needs fixing before anyone reruns it.
Cancel the OpenAI eval run evalrun_91x for our support classifier, it's using last month's dataset.
Send a big dataset to OpenAI in chunks as part of an upload, which is how files above the single-request limit get in.
Add the next part of our fine-tuning JSONL to OpenAI upload upload_55a and tell me how many parts are done.
Sends one chunk of bytes to an Upload object, used when moving large files to OpenAI in pieces.
Stops a batch job that is still running, for example after spotting a mistake in its inputs.
Ends an evaluation run in progress when its setup is wrong or results are no longer needed.
Stops a model response that is running in background mode, identified by its ID.
Halts an assistant run on a thread before it finishes, to stop unwanted tool calls or output.
OpenAI APIs support model inference, responses, assistants, files, vector stores, fine-tuning, embeddings, images, audio, moderation, batches, realtime, and administration workflows. Integrations authenticate with API keys and can build AI assistants, multimodal applications, retrieval workflows, automation agents, and content-processing systems.
Many teams build on OpenAI's API. The connection lets Claude or Codex in type.com manage those OpenAI resources, such as batches, evals, uploads, and runs, while engineers coordinate in one shared thread.
OpenAI uses an API key that you add once to the connection. Admins decide which Spaces can use it, and the key stays out of prompts and thread messages.
Only people working in Spaces that the OpenAI connection is shared with. Admins choose those Spaces, so cancelling batches, evals, or runs stays with the engineers responsible for your OpenAI usage.
Not exactly. type.com uses connectors and connections to give selected Spaces access to approved app tools and data. Some connectors use hosted MCP, while others use OAuth, API keys, service accounts, or custom APIs.
Connect Abstract API to Claude in type.com to validate emails, check VAT categories, and verify IBANs from a shared thread.
Connect Addressfinder to Claude in type.com to autocomplete, verify, and pull metadata for Australian addresses right from a team thread.
Use Addresszen with Claude in type.com to resolve US address suggestions and confirm an API key is active before running lookups.
Connect Agility CMS to Claude in type.com to fetch content items, check models, and pull sync logs from a shared thread.