Skip to content

OpenAI integrationIn development

Analyze production issues with your own OpenAI key

The OpenAI integration is in development. It is designed so you connect your organization's OpenAI API key, choose a model, and ROOKON sends each issue's gathered context to OpenAI for a structured diagnosis and a proposed fix.

The problem

Your AI account, terms and budget should stay yours

Many teams already have an OpenAI account with agreed terms, usage limits and data settings. A tool that resells AI access forces a second set of terms on you, plus an unclear markup.

  • Opaque per-seat “AI credits” on top of your subscription
  • Data flowing through a vendor's account you can't inspect
  • No control over which model is used
  • Usage limits you can't manage yourself

Live demo 1 of 3

Connect your OpenAI key safely

See what happens when a key is saved: format check, a fresh data key, AES-256-GCM encryption and a masked hint. Runs in your browser and sends nothing.

  • Use the sample key
  • Press “Encrypt and save”

rookon · settings · AI provider · OpenAI

Simulated data

Connect OpenAIIn development

Don’t paste a real key. This demo runs only in your browser and sends nothing. Use the sample key instead.

Envelope encryption

When you save a key:

  1. a fresh data key encrypts it (AES-256-GCM),
  2. a master key that never touches the database wraps the data key,
  3. only ciphertext and a masked hint are stored,
  4. the worker decrypts it for a single analysis request, then forgets it.

The API returns only maskedKey after saving, never the key itself, to anyone.

How ROOKON helps

What you get

  • Bring your own key

    Paste an OpenAI API key (ideally one created for ROOKON in a dedicated project) and analyses run on your account.

  • Choose the model per task

    Select from the models your key can access, per task: one for diagnosis, another for writing code. ROOKON validates the key and model with a minimal request when you connect.

  • Encrypted at rest, server-side only

    Keys are stored with envelope encryption (AES-256-GCM) and decrypted only inside the background worker at the moment of a request. They are never sent to the browser or written to logs.

  • Structured, reviewable output

    Responses are requested in a fixed structure (the six questions plus an optional diff), so every recommendation reads the same way.

  • Usage you can see

    Token usage is recorded per analysis, so you know what each investigation consumed. OpenAI bills that usage directly to your account.

Live demo 2 of 3

Pick an OpenAI model per task

After connecting, choose which of your models diagnoses issues and which writes the code. Model names are examples: ROOKON lists the models your own key can use.

  • Change the code generation model
  • Set review revisions back to “Same as fallback”

rookon · workspace settings · AI models

Simulated data

Connected providers

Model names below are examples. In ROOKON the list is fetched from your provider account with your key.

TaskModel

Diagnosis

A fast, inexpensive model is usually enough.

Code generation

Pick your strongest coding model. This is where quality matters most.

Review revisions

Defaults to your code generation model.

Uses Code generation: your-strongest-coding-model

Marketing copy

Defaults to your diagnosis model.

Uses Diagnosis: your-fast-model

Support replies

Defaults to your diagnosis model.

Uses Diagnosis: your-fast-model

Content writing

Defaults to your marketing copy model.

Uses Diagnosis: your-fast-model

example pricesPer-run estimates use example token prices and typical token counts. Your budget uses your real prices.

The model you pick for code generation writes every proposed change. Model routing is in development with the rest of ROOKON.

How the connection is designed to work

  1. 01

    Create a key

    Create an API key in your OpenAI account, scoped to a project you use for ROOKON.

  2. 02

    Connect

    An owner or admin adds the key in ROOKON's AI settings and picks a model.

  3. 03

    Validate

    ROOKON makes a minimal test request to confirm the key and model work.

  4. 04

    Analyze

    Issues are analyzed in background jobs, on demand or automatically, depending on your rules.

Live demo 3 of 3

Cap what OpenAI costs you

Set a monthly budget for analyses on your OpenAI key and decide what runs automatically. Prices from the model picker above feed the estimate. Simulated.

rookon · workspace settings · AI budget

Simulated data

Billed by your AI provider, on your key. ROOKON stops before the cap.

What runs automatically

Your model’s price per 1M tokens

Example prices. Check your provider’s current pricing for your model.

Simulated day 1 of 30

Live

$0.00 of $20 this month

on track

projected month-end $0.000 tokens0 automatic · 2 waiting

Detected issues (newest first)

  • infoRarely used setting causes console warningday 1
    This issue is below the severity your autopilot mode analyzes automatically.
  • infoImage without dimensions on /blogday 1
    This issue is below the severity your autopilot mode analyzes automatically.

Simulated month. Most impactful issues are analyzed first; ROOKON reserves the worst case before each call so a single analysis can’t push you over the cap.

Try: Pick a pricier diagnosis model above · Raise the cap

FAQ

Common questions

Is the OpenAI integration available now?

Not yet. It is in development, and this page will say so as soon as you can connect a key.

Who pays for OpenAI usage?

You do, directly to OpenAI under your own account. ROOKON does not resell or mark up model usage.

What data is sent to OpenAI?

Only the context needed for the issue being analyzed: privacy-filtered error details, a text summary of relevant session events and the related source snippets. Raw session recordings are not sent.

How is my API key protected?

It is encrypted at rest with a per-record data key, which is itself encrypted with a master key kept outside the database. Only the background worker decrypts it, at the moment of a request, and it can be replaced or deleted at any time.

Can I use another OpenAI-compatible endpoint?

Support for OpenAI-compatible endpoints is planned as a separate option, so self-hosted or alternative providers can be connected in the same way.

Put your OpenAI account to work on real issues

Sign up and connect your OpenAI key in the setup journey.