FieldCampAgents in Workflows | FieldCamp
Use AI agents as intelligent steps inside FieldCamp workflow automations. Combine triggers and conditions with AI-powered actions for smarter business processes.
FieldCamp's workflow automations let you set up triggers, conditions, and actions that run automatically. Adding an AI Agent as one of those actions makes your workflows smarter — instead of just sending a templated email, the agent can draft a personalized message, decide what to say based on context, or handle a conversation.
What It Means
A regular workflow action is fixed: "send this email template" or "change the status to Complete." An AI Agent action is intelligent: "look at this job, figure out the right follow-up message, and draft it." The agent brings reasoning and personalization to your automations.
How to Add an Agent to a Workflow
- Go to Settings > Workflow Automation or open the Workflow Builder
- Create a new workflow or edit an existing one
- Set your trigger (the event that starts the workflow, like "Job completed")
- Add any conditions you need (like "Only if job value is over $200")
- Drag a Do something action node onto the canvas
- In What should happen?, choose AI Agent
- Write the AI Prompt and choose the AI Model. Data Source Entity is optional -- set it when the AI should read a record such as the job or the client
- Save and activate the workflow
The workflow builder has no separate "AI Agent" node in the palette, and you do not pick one of your saved agents here. The AI Agent action is its own prompt: when the trigger fires and conditions are met, FieldCamp reads the data source, runs your prompt, and returns text that later steps can use.
The AI Agent workflow action only produces text. It cannot place calls or send messages by itself — chain a Send Email, Send SMS, or Send WhatsApp action after it to deliver what it wrote.
Real-World Examples
Follow-Up Email After Job Completion
Trigger: A job status changes to "Completed"
Condition: The job value is over $200
AI Agent action: The Follow-Up Agent reviews the job details, drafts a personalized thank-you email mentioning the specific service performed, and includes a link to leave a review.
Next action: The Send Email action delivers the agent's draft to the client.
Without the agent, you would send the same generic "Thank you for your business" email to every client. With the agent, each email references the actual work done and feels personal.
Appointment Confirmation Message
Trigger: A visit is scheduled for tomorrow
AI Agent action: The AI Agent step reads the visit and drafts a short confirmation message with the date, time, and technician name.
Next action: A Send SMS or Send WhatsApp action delivers the drafted message to the client.
Result: Fewer no-shows because every client gets a personal confirmation the day before.
Overdue Invoice Escalation
Trigger: An invoice is 14 days past due
Condition: No payment has been recorded
AI Agent action: The AI Agent step drafts a polite but firm reminder with the outstanding amount and due date.
Next action: A Send Email action delivers it. If it is still unpaid after 7 more days, a second workflow escalates with a stronger message.
New Request Triage
Trigger: A new request is created
AI Agent action: The AI Agent step reads the request description and returns a category, an urgency level, and a short recommendation.
Next action: An Update Request Stage or Update Field action writes that result onto the request.
What Data Can You Pass to the Agent?
When you configure the AI Agent action you can set a Data Source Entity, and you can insert fields from the trigger event straight into the prompt with the variable picker. Common fields include:
| Data | Use Case |
|---|---|
| Client name and contact info | Personalize emails and calls |
| Job or visit details | Reference specific work performed |
| Invoice amount and due date | Payment reminders and follow-ups |
| Request description | Triage and categorization |
| Technician name | Include in client communications |
| Scheduled date and time | Appointment confirmations |
The AI Agent action reads the data source entity you selected and the variables you inserted, then answers your prompt.
Tips for Agents in Workflows
- Keep the agent focused — the workflow handles the trigger and conditions, so the agent only needs to handle its one action
- Test the prompt on safe data first — run the workflow against a test record before pointing it at real customers
- Chain agents with other actions — an agent can draft a message, and a separate Send Email action can deliver it. This lets you review agent outputs before they go out if needed
- Monitor execution history — check the workflow dashboard to see how often the agent runs and whether it succeeds
The workflow itself must be activated for the AI Agent step to run. The Active / Paused / Draft status on the Agents page applies to standalone agents, not to this workflow action.
Start with a simple combination: trigger on job completion, have the agent draft a personalized follow-up email, and send it automatically. Once you see how it works, build more complex flows.
Next Steps
- Workflow Automation Overview — understand triggers, conditions, and actions
- Building Workflows — step-by-step guide to the workflow builder
- Workflow Examples — real-world automations you can copy
- Creating & Configuring Agents — set up the agent before adding it to a workflow
- AI Agents Overview — understand what agents are and how they work