Agentic Actions Across Integrations

When building an AI agent application, you’ll likely need to give it the ability to perform read, write, and update operations in your customers’ third-party applications like CRMs, calendar apps, ticketing platforms.

Paragon makes it easy to equip your AI agents with third-party actions, and there are two methods for doing so:

1. Workflows with Paragon-managed steps for common operations in a third-party provider

2. Proxy API where Paragon manages all auth-related details for every third-party endpoint

Workflows

Building workflows for agentic actions is very simple, especially if you just need to perform a simple write operation like creating a ticket in Intercom.

Where workflows really shine is when the action required isn’t a simple third-party API call. For example, if you need to get a user’s Google calendar availability, your AI agent can use a workflow like this one to calculate the availability based on the existing events that the Google Calendar API returns.

In your code, you’ll trigger the workflows via POST requests to give your AI agent the ability to work with the third-party integration provider, whether it’s to read, write, or delete data.

const getSdrSchedule = FunctionTool.from(
      async ({ dummy }: {dummy: string}) => {
          console.log("Getting SDR Calendar");
        const response = await getGoogleCalendarAvailability(signJwt(userId));
        console.log(response);
        if (response.busy) {
          return response.busy;
        }
        return "Calendar could not be pulled";
      },
      {
        name: "getSdrSchedule",
        description: "Use this function when a user asks to schedule a meeting with the SDR. This function returns when " +
            "the SDR is busy and UNABLE to meet." +
            "Times are provided in UTC format. Use the convertUtcDatetimeToPstDatetime tool to convert UTC datetimes to" +
            "PST before returning results.",
        parameters: {
          type: "object",
          properties: {
            dummy: {
              type: "string",
              description: "dummy",
            },
          },
          required: [],
        },
      },
    );

export async function getGoogleCalendarAvailability(jwt: string) {
    return await fetch(process.env.GET_GOOGLE_CALENDAR_EVENTS_ENDPOINT ?? "", {
        method: "POST",
        headers: {
            "Content-Type": "application/json",
            Authorization: "bearer " + jwt,
        },
    })
        .then((response) => response.json())
        .catch((error) =>
            console.log("Error getting Google Calendar events: " + error)
        );
}

Enabling your AI agent with both create and read workflows, you can create interactions like this one where your AI application truly be assistant-like.

Proxy API

The second way Paragon can help build your agentic action use case is to use the Paragon SDK and Proxy API. These proxy API endpoints handle the authentication mechanism, such as inheriting the auth for a user in Slack to send a message as the user.

Calling Paragon’s proxy API is a much simpler developer experience and can be used in a function tool by your AI agent, like in the code example above.

await paragon.request('slack', '/chat.postMessage', {
	method: 'POST',
	body: {
		channel: 'CXXXXXXX0' // Channel ID,
		text: 'This message was sent with Paragon Connect :exploding_head:'
	}
});

Wrapping Up

Paragon’s platform provides different ways AI applications can perform agentic actions in a third-party platform. Our team is constantly thinking of ways we can introduce features in our iPaaS product to empower our customers to build out integrations for their AI use cases. To learn more about how Paragon can enable your AI application to perform actions across integrations and our exciting new AI-focused features, book a demo to schedule some time with our team.