Stop digging through docs to find OAuth details. This setup uses an AI model to suggest the authorization URL, token URL, audience, and a clean service name from a simple input. It helps engineers and IT teams speed up API onboarding and cut guesswork.

The flow starts when another workflow calls it with a short JSON payload. The LLM Bus prepares a prompt and sends it to the OpenRouter Chat Model using a Llama 3.3 based model. A Structured Output Parser shapes the AI reply into a clear JSON schema. A Code step then conforms the fields to fit your system. Each result includes a confidence score so you can set quality thresholds and route low confidence results for review.

Run it as a service inside your n8n stack. Provide an OpenRouter API key and pass fields like name and audience. Expect faster setup, fewer mistakes, and consistent output. Use it to accelerate API onboarding, internal tooling, and sandbox testing where OAuth settings are often missing or unclear.

• Runs when executed by another workflow so it acts like a reusable service.
• Uses the OpenRouter Chat Model with tuned options for balanced quality and cost.
• LLM Bus organizes the prompt and routes the model and the output parser.
• Structured Output Parser enforces a JSON schema for stable fields and types.
• Code step conforms field names and structure to match your app needs.
• Confidence score on every result to control routing and human review.
• Retry and token limits set on the model to reduce failures and timeouts.

• Reduce manual research from 2 hours to 5 minutes
• Automate 80 percent of repetitive OAuth lookups
• Improve data accuracy by 30 percent with schema checks
• Handle 10 times more API onboardings each week
• Connect AI results to any internal workflow with one call

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with OpenRouter. See the Tools Required section above for links to create accounts with these services.
3. Get your OpenRouter API key from the provider dashboard under the API section.
4. In n8n, double click the OpenRouter Chat Model node. In the Credential to connect with dropdown, click Create new credential and follow the on screen instructions to add your OpenRouter API key. Give the credential a clear name.
5. Open the LLM Bus node and review the prompt text. Add examples or naming rules that match your APIs to improve accuracy.
6. Open the Structured Output Parser node and check the JSON schema. Include fields for service name, audience, authorization URL, token URL, details, and confidence if you plan to store them.
7. Open the Code step named Conform JSON and adjust key names to match your downstream system or database fields.
8. Test the service: In the When Executed by Another Workflow trigger, use the sample JSON with name and audience, then execute. Verify the parser output and the conformed JSON.
9. Call this service from another workflow using the Execute Workflow node. Pass a JSON body like { "name": "Atlassian", "audience": "api.atlassian.com" } and review the returned fields.
10. Troubleshoot parsing issues by lowering temperature in the model options, checking response format is JSON object, and raising max tokens if outputs are cut off. If confidence is low, route to manual review or retry with a more specific prompt.