Learn how to browse, build, and manage AI agents in FieldCamp to automate field service workflows from one central marketplace.

The AI Agents marketplace in FieldCamp is where you browse, build, and manage AI agents that automate repetitive work across your field service business. Whether you need an agent to follow up on quotes, confirm appointments, or chase overdue invoices, the marketplace gives you a single place to deploy and tune them. This guide walks you through navigating the marketplace, configuring an AI agent for your team, and keeping it running smoothly day to day.

What the AI Agents marketplace is

The marketplace is a curated library of pre-built AI agent templates plus the tools you need to create your own from scratch. Each AI agent is a focused assistant with its own instructions, connected tools, and on/off switch — designed to handle one job well rather than try to do everything.

You can think of agents as digital teammates that sit alongside your dispatchers, technicians, and office staff. They never sleep, never forget a task, and they follow the playbooks you set. The marketplace is where you decide which ones to hire.

AI agents work best when paired with FieldCamp's broader automation toolkit. If you're new to the platform, start with the FieldCamp workflow from lead to payment to see where agents fit.

How to access the marketplace

The marketplace lives inside the Agents section of your sidebar. From there you can browse templates by category, view active agents, or start a blank agent. If you're not sure where to find it, the navigating FieldCamp interface and sidebar guide shows you every section of the app.

You can also reach the marketplace from the FieldCamp Command Centre — just ask the assistant to "show me available AI agents" and it will surface relevant templates.

Browsing agent templates

Templates are grouped into categories so you can find what you need quickly:

Communication agents — handle calls, texts, and emails for after-hours coverage or auto-replies
Scheduling agents — book appointments, reschedule visits, and confirm arrival windows
Financial agents — chase deposits, send invoice reminders, and reconcile payments
Reporting agents — generate weekly performance summaries and revenue digests
Operations agents — keep job statuses up to date and notify your team about exceptions
Research agents — look up client history, check service areas, and answer FAQs

Start with one template that solves your biggest daily headache — usually after-hours calls or invoice follow-ups — and expand from there.

Building an AI agent step by step

Choose a template or start from scratch

Open the marketplace from the Agents page. Pick a template that closely matches the task you want to automate, or click New Agent to start blank. Templates give you a head start by pre-filling instructions and recommended tools. For a deeper walk-through, see creating and configuring agents.

Name and describe the agent

Give the agent a clear name like "After-Hours Booking" or "Invoice Reminder 30 Days." A short description helps your team understand what the agent does at a glance.

Write the agent's instructions

Tell the agent exactly what to do, in plain language. Describe the goal, the tone of voice, what data to look up, and what to do when something is unclear. Be specific — vague instructions lead to inconsistent results.

Connect the right tools

Agents need permission to read or write data — pulling a client's invoice, sending an SMS, creating a calendar event. Connect only the tools the agent truly needs. See the full list of available connections in the agent connections reference, and review MCP connections and custom triggers for advanced integrations.

Test the agent

Run a few test scenarios before going live. Send a mock job, place a test call, or trigger a sample workflow. Tune the instructions until the agent behaves the way you'd want a new hire to behave.

Activate and monitor

Flip the agent to Active. Check the activity log over the first few days to confirm it's handling real cases correctly. You can pause an agent at any time without losing its configuration.

Managing agent status and permissions

Every agent has one of three states: Active, Paused, or Draft. Active agents run continuously, paused agents keep their configuration but stop working, and drafts let you stage changes safely.

Because agents act on behalf of your business, control who can build and edit them. Review your roles and permissions to make sure only trusted team members can change live agents. Most businesses limit agent editing to owners and operations managers while giving everyone else read-only visibility.

Agents with access to financial tools — like sending invoices or charging cards — should always require a manager-level role to edit. Mistakes in instructions can result in customer-facing errors.

How agents fit with the rest of FieldCamp

AI agents are most powerful when they work alongside your existing automations and dispatching:

AI Dispatcher

Let the AI Dispatcher assign jobs while a follow-up agent confirms with the customer.

Workflow builder

Trigger agents inside workflows — for example, "send an SMS, wait a day, then call."

Inbox

Have an agent triage incoming emails or draft replies before a human reviews them.

Command Centre

Ask the assistant to build, pause, or summarize agents using natural language.

Voice-based scenarios are a great starting point too — voice agents can answer after-hours calls and book appointments by phone.

Monitoring agent performance

Open any agent to see its recent activity, success rate, and any errors it hit. Look for these signs that an agent needs tuning:

Repeated handoffs to a human for the same scenario — add that case to the instructions
Customer confusion in transcripts — simplify the tone or clarify what the agent should ask
Missing data errors — confirm the agent has the right tool connections and permissions

For deeper insights, build a dashboard in FieldCamp analytics that tracks agent volume, resolution rate, and impact on jobs booked or invoices collected. You can also use custom dashboards and boards to assemble role-specific views for owners and operations managers.

Common patterns to copy

A few proven setups other field service teams run from day one:

After-hours booking agent — captures requests overnight and creates a job in FieldCamp by morning
Quote follow-up agent — texts prospects 2 and 7 days after sending an estimate
Deposit chaser — politely asks for a deposit once a job is scheduled
Post-job satisfaction agent — calls the customer the day after the visit and logs the rating
Weekly digest agent — emails owners a Monday summary of jobs, revenue, and exceptions

When you build an agent that touches money, double-check your tax and payment settings first. The tax settings guide covers the basics.

Troubleshooting

If an AI agent isn't behaving the way you expect, work through these steps:

Re-read the instructions out loud — would a new employee understand them?
Check the connected tools — is the agent missing read or write access to something it needs?
Look at the activity log for the exact moment things went wrong
Pause the agent and revert to the last working configuration while you debug

For broader help across the platform, the troubleshooting common issues and fixes guide is a good starting point.

Building and Managing AI Agents in FieldCamp