OpenAI connector for AI agents

Tools your LLM ops agent reaches for on OpenAI, scoped per user.

CALL ANY TOOL

62 tools for LLM ops in OpenAI: create and track batch jobs, run evals, manage fine-tuning jobs, build and search vector stores, manage conversations and code interpreter containers, and read stored responses, completions, and files.

openai_batch_cancel

Cancel batch

Cancel an in-progress batch job. Returns the batch with status cancelling; it moves to cancelled within about 10 minutes, and any partial results stay in its output file. Use this to stop a batch that is no longer needed. Use openai_batch_get to watch it reach cancelled.

Parameters

Name

Type

Required

Description

batch_id

string

Required

ID of the batch to cancel. Get it from openai_batch_list. Example: batch_abc123.

openai_batch_create

Create batch

openai_batch_get

Get batch

openai_batch_list

List batch

openai_chat_completion_message_list

List chat completion message

openai_chat_completion_update

Update chat completion

openai_container_file_content_get

Get container file content

openai_container_file_create

Create container file

openai_container_get

Get container

openai_container_list

List container

openai_conversation_item_create

Create conversation item

openai_conversation_item_get

Get conversation item

openai_eval_list

List eval

openai_eval_run_create

Create eval run

openai_eval_run_get

Get eval run

openai_eval_run_output_item_list

List eval run output item

openai_eval_update

Update eval

openai_file_content_get

Get file content

openai_fine_tuning_job_cancel

Cancel fine tuning job

openai_fine_tuning_job_checkpoint_list

List fine tuning job checkpoint

openai_fine_tuning_job_get

Get fine tuning job

openai_fine_tuning_job_list

List fine tuning job

openai_fine_tuning_job_pause

Pause fine tuning job

openai_fine_tuning_job_resume

Resume fine tuning job

openai_response_get

Get response

openai_response_input_item_list

List response input item

openai_vector_store_create

Create vector store

openai_vector_store_file_content_get

Get vector store file content

openai_vector_store_file_list

List vector store file

openai_vector_store_search

Search vector store

For more tools, view docs.

Build your Agent

Same auth pattern across LangChain, OpenAI, Anthropic, and Google ADK.

import { ScalekitClient } from "@scalekit-sdk/node";
import { createReactAgent } from "@langchain/langgraph/prebuilt";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);

// OpenAI tools scoped to this user
const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["openai"], toolNames: [
    "openai_vector_store_create",
    "openai_vector_store_file_create",
    "openai_vector_store_search"] },
  pageSize: 100,
});

const agent = createReactAgent({ llm, tools });
await agent.invoke({ messages: [{ role: "user", content: "Build a Help Center vector store from these files and search it for our refund policy" }] });

import OpenAI from "openai";
import { ScalekitClient } from "@scalekit-sdk/node";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);
const openai = new OpenAI();

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["openai"] }, pageSize: 100,
});

const res = await openai.chat.completions.create({
  model: "gpt-5",
  messages: [{ role: "user", content: "Build a Help Center vector store from these files and search it for our refund policy" }],
  tools,
});

// Execute the tool call with the user's vaulted OpenAI access token
await sk.tools.executeTool(res.choices[0].message.tool_calls[0], "user_123");

import Anthropic from "@anthropic-ai/sdk";
import { ScalekitClient } from "@scalekit-sdk/node";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);
const anthropic = new Anthropic();

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["openai"] }, pageSize: 100,
});

const msg = await anthropic.messages.create({
  model: "claude-sonnet-5",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Build a Help Center vector store from these files and search it for our refund policy" }],
  tools,
});

// Tool call runs with the user's vaulted OpenAI access token
await sk.tools.executeTool(msg.content, "user_123");

import { Agent } from "@google/adk/agents";
import { ScalekitClient } from "@scalekit-sdk/node";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["openai"] }, pageSize: 100,
});

const agent = new Agent({
  name: "llm_ops_agent",
  model: "gemini-2.5-pro",
  instruction: "OpenAI tools scoped to this user",
  tools,
});

await agent.run("Build a Help Center vector store from these files and search it for our refund policy");

Try these prompts

Copy any prompt into your agent. Each maps directly to a OpenAI tool. Click to copy, paste into your agent, done.

Batches and fine-tuning

Copy the prompt

Copied

Start a batch job from the uploaded requests file on the chat completions endpoint.

Copy the prompt

Copied

Show the status and request counts of yesterday's batch.

Copy the prompt

Copied

Pause the fine-tuning job that is running now.

Evals

Copy the prompt

Copied

List our evals and their testing criteria.

Copy the prompt

Copied

Start a new run of the support answers eval.

Copy the prompt

Copied

Show the failing items from the last eval run.

Vector stores and history

Copy the prompt

Copied

Create a vector store called Help Center from these uploaded files.

Copy the prompt

Copied

Search the Help Center vector store for our refund policy.

Copy the prompt

Copied

List the stored chat completions from last week.

SEE HOW AUTH WORKS

Each user connects their own OpenAI access token once, and Scalekit sends it with every call. Access tokens stay vaulted, every call is scope checked, and every action is logged.

Authorize

Your user connects

OpenAI

once. We tie it to their identity and the meetings they approved — no shared bot account, no org-wide access

Who:

user ‘A’

when:

Once per user

access:

Limited to user

Store

Their

OpenAI

token lives in a vault scoped to them. User A's meetings are never reachable by an agent acting for user B, even on the same connection

vault:

encrypted

scope:

per-user

tokens:

auto-refreshed

Resolve

When your agent calls a

OpenAI

tool, we fetch the right token server-side. It never touches your agent, never appears in the LLM context, never shows up in your logs

speed:

~40ms

check:

before every call

seen by:

nobody

Audit

Every

OpenAI

tool call is logged — who triggered it, which meeting was fetched, what came back. 90 days of history, tied to the user who authorized it

history:

90 days

export:

SIEM-ready

logged:

every call

Test other agents

See the same per-user auth pattern across other AI and LLM connectors.

Engineering Teams

DevOps assistant agent

Polls GitHub for failing checks and stale PRs, opens Linear issues for the ones that need work, and posts a daily digest to Slack. It acts as the engineer, not a shared service account.

Engineering Teams

Engineering standup agent

Pulls commits from GitHub and GitLab, tracks issue movement in Jira, and posts a per-engineer standup brief to Slack. Each engineer's activity is read on their own delegated OAuth.

Why Scalekit

Secure your agent's access. Connectors ship in minutes

01.

Shared tokens break per-user analytics

A shared OpenAI access token looks fine in a demo. In production every batch job and fine-tune looks like one service account, and you cannot tell which user triggered it. Scalekit resolves the credential of the actual user who triggered the agent, never a shared bot.

// shared key
audit → bot_service_account

// scalekit
audit → user_abc ✓

02.

Authentication is not authorization

03.

Multi-tenancy is architectural

04.

OpenAI today. Ten connectors tomorrow.

“Our agents act across Salesforce, Gong, Google Drive, and more, on behalf of every customer. Scalekit behind the scenes meant we can keep adding tools without ever rebuilding how credentials or tool calling work.”

Venu Madhav Kattagoni

Head of Engineering / Von

FAQs

Frequently Asked Questions

Does the agent access OpenAI as the user or through a shared access token?

As the user. Each user connects their own OpenAI access token once, and Scalekit sends it with every call. Audit logs attribute every batch job, eval, and fine-tune to that user, not a shared service account.

Where is the OpenAI access token stored?

In Scalekit's managed AES-256 token vault, namespaced per tenant. To rotate it, the user reconnects with a new access token. Revocation is a single dashboard action. Credentials never appear in prompts, logs, or LLM context.

Can I limit what the agent does in OpenAI?

Yes. Filter by tool name in listScopedTools to expose only what you want, for example the get and list tools for batches, evals, and vector stores without openai_batch_create or openai_fine_tuning_job_create, which start new jobs. No OpenAI tool deletes data. Scalekit also enforces scope checks before every API call.

What happens when a user revokes OpenAI access?

The connection is invalidated on the next tool call. Subsequent requests for that user fail closed with a clear error. Other users in the tenant remain unaffected. The event is logged for audit.

What can the agent read from past OpenAI calls?

Stored responses with openai_response_get and their input items, Chat Completions stored with store set to true through openai_chat_completion_list, and conversation items. openai_response_input_token_count counts the input tokens a Responses request would use without generating one.