Quickstart

Five minutes from zero to your AI agent reading and writing your UserPulse boards.

1Create an API key

Go to Settings → API keys and click New key. Pick:

Scope: account (all your boards) or board (one board only).
Permission: read or read + write.

Copy the secret up_… immediately. You won't see it again.

2Make a request

curl

curl -H "Authorization: Bearer up_..." \
  https://your-deployment.convex.site/api/v1/boards

javascript

const res = await fetch(
  "https://your-deployment.convex.site/api/v1/boards",
  { headers: { Authorization: `Bearer ${process.env.USERPULSE_KEY}` } },
);
const { boards } = await res.json();

python

import os, requests

r = requests.get(
    "https://your-deployment.convex.site/api/v1/boards",
    headers={"Authorization": f"Bearer {os.environ['USERPULSE_KEY']}"},
)
print(r.json()["boards"])

3File feedback as the agent

curl

curl -X POST \
  -H "Authorization: Bearer up_..." \
  -H "Content-Type: application/json" \
  -d '{"title":"Add CSV export","description":"Customers want it"}' \
  https://your-deployment.convex.site/api/v1/boards/product/feedback

javascript

await fetch(
  "https://your-deployment.convex.site/api/v1/boards/product/feedback",
  {
    method: "POST",
    headers: {
      Authorization: `Bearer ${process.env.USERPULSE_KEY}`,
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      title: "Add CSV export",
      description: "Customers want it",
    }),
  },
);

python

requests.post(
    "https://your-deployment.convex.site/api/v1/boards/product/feedback",
    headers={
        "Authorization": f"Bearer {os.environ['USERPULSE_KEY']}",
        "Content-Type": "application/json",
    },
    json={"title": "Add CSV export", "description": "Customers want it"},
)

4Read or update an item

See the full request/response shapes in the interactive API reference.

5Plug it into your agent

Read the AI Agents Guide for ready-to-paste tool definitions for Claude and GPT.