Connect Claude to Observa
Follow these steps to give your AI assistant access to real-world observations. You'll be up and running in about 5 minutes.
You'll need:
- A Claude Desktop app (or any MCP-compatible client)
- An Observa account (free to create)
Create your Observa account
- Go to app.observa.com and click “Try for Free”.
- Sign in with Google or GitHub — no separate password needed.
- Your account is created automatically on first sign-in.
Choose your plan
Observa offers two ways to use the platform:
Category Intelligence subscription
For ongoing competitive intelligence. Subscribe to a category and tier from the dashboard. Category Intelligence plans start at $249/month.
Pay-as-you-go observations
Order observations on-demand, pay-as-you-go. Just add a payment method — no subscription required.
A free Category Intelligence tier is available to explore the platform.
Generate an API key
- Navigate to API Keys in the dashboard sidebar.
- Click “Generate API Key”.
- Give it a label (e.g., Claude Desktop).
Important
Copy the key immediately — it is only shown once. The key looks like: obs_a1b2c3d4...
Configure Claude Desktop
- Open Claude Desktop → Settings → MCP Servers (or Developer → Edit Config).
- Add the Observa server configuration:
{
"mcpServers": {
"observa": {
"url": "https://gpu-worker.tailc90307.ts.net/api/mcp/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}Replace YOUR_API_KEY with the key you copied in Step 3.
Save the file and restart Claude Desktop.
Place your first observation
Open a new conversation in Claude Desktop and try these prompts:
“What observation types does Observa offer?”
Claude will use the describe_services tool.
“Order a billboard observation for Coca-Cola at Times Square, New York”
Claude will use the order_billboard_observation tool.
If it works, you'll see a confirmation with the observation ID and cost.
Other MCP Clients
Any MCP-compatible client can connect to Observa using the same configuration. Add this to your client's MCP server settings:
{
"mcpServers": {
"observa": {
"url": "https://gpu-worker.tailc90307.ts.net/api/mcp/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}Refer to your MCP client's documentation for where to add server configurations.
Troubleshooting
Connection failed?
Check that your API key is correct and has not been revoked. You can verify active keys in the API Keys section of your dashboard.
Tool calls blocked?
Ensure you have an active subscription or a payment method on file. Visit payment setup to add one.
Need help?
Return to your dashboard to review your account status, API keys, and subscription details.
Start your first observation
Create your account and connect your AI assistant to real-world observations in minutes.
Create your account