Skip to main content
Your Spark agent can connect to any tool or service that has an API — not just the ones listed on this page. The built-in integrations below are one-click setup, but your agent isn’t limited to them.
Your agent can work with any software. If a tool has an API, OAuth, webhooks, or any kind of programmatic access, your agent can use it. Just provide the credentials (API key, access token, OAuth connection) and your agent figures out the rest. The 30+ built-in integrations are just the ones with pre-configured setup — your agent’s capabilities are unlimited.

Built-In vs. Custom Integrations

The key insight: your agent is a general-purpose operator, not a pre-built connector. It can read API documentation, authenticate with credentials, make HTTP requests, and interpret responses. This means the integrations page is a starting point, not a ceiling.

Your Agent Doesn’t Just Read — It Acts

This is critical to understand: your agent doesn’t just pull data from these platforms. It performs actions on them. You talk to your agent, and it goes and does the work across whatever tools you’ve connected.
“Update the Q3 revenue forecast in Google Sheets to reflect the new deal we just closed” “Create a Jira ticket for the checkout bug the customer reported, assign it to the engineering team, and mark it as high priority” “Review the latest PR on GitHub, check if the tests pass, and leave a comment with your analysis” “Send an email to the client confirming their appointment, then add it to Google Calendar and create a follow-up task in Asana for next week” “Check our Stripe dashboard, find all customers whose subscription failed this month, and draft a re-engagement email for each one” “Add a new row to our Airtable tracking sheet with the vendor’s quote details, then update the Notion procurement page with the status”
Your agent is an operator across your entire tool stack. It doesn’t matter if the work needs to happen in Gmail, Jira, Stripe, Google Sheets, or Slack — you talk to one agent, and it handles everything across all your connected platforms. No switching tabs, no manual data entry, no copy-pasting between tools.

How Built-In Integrations Work

  1. Go to Integrations in your workspace sidebar
  2. Find the tool you want to connect
  3. Add your API key or credentials
  4. Your agent automatically gets access on its next restart
That’s it. No code, no configuration files.

Available Integrations

💻 Development

🎨 Design

📋 Productivity

💬 Communication

🗄️ Data & Payments

📈 Marketing

🛒 Commerce

Setting Up an Integration

1

Navigate to Integrations

Go to your workspace and click Integrations in the sidebar.
2

Find Your Tool

Browse by category or search for the integration you need.
3

Add Credentials

Each integration needs different credentials — typically an API key or access token. Click Connect and fill in the required fields.
4

Agent Restart

Your agent picks up new integrations on restart. You can trigger a restart from the agent’s detail page, or it happens automatically.
Not sure where to find your API key? Each integration’s connect dialog includes a link to the service’s documentation where you can generate credentials.

OAuth Integrations

Some integrations use OAuth — meaning you sign in with your existing account instead of copying API keys:
  • Gmail — sign in with your Google account to grant email access
  • Google Calendar — sign in to grant calendar read/write access
  • Outlook — sign in with Microsoft to grant email and calendar access (see Outlook)
  • Linear — sign in with your Linear workspace; no API key needed
OAuth integrations are more secure because your password is never shared with Spark. You authorize specific permissions, and you can revoke access anytime from the provider’s account settings.
If you connect both Google and Microsoft, your workspace’s Email and Calendar tabs use the most recently connected provider.

What Happens After You Connect

Once an integration is connected:
  1. Your agent knows about it — it can see what tools are available in its configuration
  2. You can ask it to use them — “Check my Gmail for unread emails” or “Create a GitHub issue for this bug”
  3. Scheduled tasks can use them — “Every morning, check Slack for any messages I missed overnight”
  4. Automations can use them — webhook triggers can instruct the agent to use connected integrations as part of processing
Integration credentials are encrypted and stored securely. Your agent accesses them at runtime — they’re never exposed in chat or logs.

Can’t Find What You Need?

If you need an integration that isn’t listed, your agent can still interact with most services using their API directly. Just provide the API documentation in a knowledge file and your agent can make HTTP requests to any service. You can also connect external services through webhooks — see the Webhooks guide for details on receiving data from any service that supports webhook notifications.