Quickstart

Run your first prompt in a few minutes. You’ll create a project and an API key, author a prompt, and call it — first with curl, then with the SDK.

1. Create a project and an API key

  1. Sign in to the app at vernaone.vernalabs.net and create (or open) a project.
  2. Go to Project settings → API keys and create a key. Copy the token — it starts with plp_ and is shown only once.
  3. Note two things you’ll need:
    • your project id (visible in project settings), and
    • the Router URL: https://router.promptlab.vernalabs.net

Keep the plp_… key secret — it’s a server-side credential. See Authentication.

2. Author a prompt

In Prompt Studio, create a prompt — for example named summarize-ticket — with a templated input {{ticket}}, then activate a version. Your prompt is now a named, versioned endpoint you can call by name.

3. Call it

With curl

curl -X POST \
  https://router.promptlab.vernalabs.net/v1/YOUR_PROJECT_ID/prompts/summarize-ticket/execute \
  -H "x-api-key: plp_xxx_xxx" \
  -H "Content-Type: application/json" \
  -d '{ "inputs": { "ticket": "User cannot log in after a password reset." } }'

With the TypeScript SDK

import { PromptLabRouterClient } from '@verna/verna-promptlab-sdk';

const client = new PromptLabRouterClient({
  routerUrl: 'https://router.promptlab.vernalabs.net',
  projectId: 'YOUR_PROJECT_ID',
  apiKey: process.env.VERNA_API_KEY, // plp_xxx_xxx
});

const res = await client.executePrompt('summarize-ticket', {
  inputs: { ticket: 'User cannot log in after a password reset.' },
});

console.log(res.outputs[0]?.text);

See TypeScript / JavaScript SDK for installation and full options.

What you get back

Every execution returns the same response envelope — outputs, the provider and model that answered, usage, cost, and timing — whether it ran on OpenAI, Anthropic, Google, or a media model.

Next steps