Activepieces Client Agent Deployment (Delivery)

Activepieces Client Agent Deployment (Delivery)

1. 01

   Create a new project in the Activepieces dashboard

   Navigate to the Projects section and click 'New Project'. Name it after the client and set the appropriate team permissions (SSO/RBAC if enabled).

2. 02

   Configure the AI agent using the drag-and-drop builder

   In the Flows tab, click 'New Flow' and select 'AI Agent'. Define the agent's prompt, tools (e.g., Gmail, Slack, HubSpot), and set human approval steps for critical actions.

3. 03

   Connect client apps via the 756+ native integrations

   Use the 'App Connections' panel to authenticate each app (e.g., OAuth for Gmail, API key for HubSpot). Test each connection with a sample trigger.

4. 04

   Set up the first automated workflow with a trigger and actions

   Choose a trigger (e.g., 'New Email in Gmail') and add actions like 'Create Lead in HubSpot' and 'Send Slack Notification'. Use the built-in test mode to validate the flow.

5. 05

   Enable audit logging and RBAC for client governance

   Go to Settings > Audit Logs and toggle logging. Under Team Management, assign roles (Admin, Editor, Viewer) to client stakeholders as needed.

6. 06

   Deploy the agent to production and monitor credits usage

   Publish the flow from the editor. In the Billing section, set a credit limit alert to avoid overage charges. For self-hosted clients, provide the Docker deployment instructions.

7. 07

   Hand off with a runbook and client credentials

   Export the flow as a JSON template from the Flows page. Provide the client with their login URL (cloud or self-hosted) and a one-page guide covering agent triggers and approval steps.