Fieldproxy MCP
The field service MCP server for Claude, ChatGPT and Copilot
The only field service MCP server where your AI can change the software itself. Ask Claude, ChatGPT or Copilot for a new field, screen or automation, and it is built in a sandbox and goes live when you approve. It runs the day-to-day too: dispatch, follow-ups, proposals.
Reply to Claude...
Illustration of Fieldproxy's MCP tools inside Claude, ChatGPT and Copilot. Names and data are examples.
Short answer
What is Fieldproxy's MCP server?
Fieldproxy's MCP server is an official Model Context Protocol connection that lets Claude, ChatGPT, Microsoft 365 Copilot and Teams, GitHub Copilot, Cursor and Codex read and act on a Fieldproxy field service workspace. It has 145 built-in tools for jobs, dispatch, customers, invoices, quotes, comms, photos and building apps. Every write, send or action is proposed first and saved only after a person confirms, and writes can be undone.
The part nobody else does
Your AI can change the software, not just the data
Every other field service connector we checked reads records and, at most, edits a few of them. Fieldproxy's MCP server can also build: new fields, screens, tables, automations and agents, made in a sandbox copy of your workspace and live only when you approve.
Add a field
Build a screen
Wire an automation
Create an agent
Add your own tools
Roll back anything
Read more: customise field service software with AI · build it with Claude or ChatGPT instead?
Connect your AI
Works in the assistant your team already uses
Connect Claude
Connect ChatGPT
Connect Microsoft Copilot
Also: Claude Code, GitHub Copilot in VS Code, Cursor and OpenAI Codex.
See your AI run your own jobs, invoices and dispatch.
Get DemoWhat your AI can do
145 tools, across the whole field operation
Not just lookups. The assistant can plan the schedule, send the follow-up and build the screen, and every change waits for a person to confirm.
Read
Customers, jobs, technicians, invoices, quotes and contracts. Accounts-receivable ageing, technician performance, funnels, and read-only SQL against your own schema.
Dispatch and maps
Spread jobs across technicians, plan daily routes and apply the schedule. Drive times, the nearest technician, live locations and job sign-off links.
Comms
Search the inbox and read threads. Draft emails and SMS from templates, then send after you confirm. Comments, notifications, email follow-ups and phone calls.
Sales documents
Preview and issue proposals, generate estimates and PDF documents.
Files and photos
Read a record's PDFs, Word, Excel and email files, and look at the photos themselves, for example to compare a job's before and after.
Monitoring
Dashboards, alerts, scheduled reports, the activity feed and usage.
Automation operations
Run a rule, retry or dismiss a failed run, and decide approvals (named approvers only).
Write, with undo
Every change is proposed, confirmed by a person, and can be undone. CSV import and export, record history and restoring deleted records.
Build
Apps, tables, automations, voice and text agents, and document templates, built as drafts in a sandbox copy of your workspace and promoted to live when you approve.
Use cases
Everything your team can do from the chat
Customise the software
Bulk updates
Imports and data migration
Dispatch and scheduling
Route planning
Billing and collections
Quotes and proposals
Customer communications
Inbox triage
Photos and documents
Daily briefing and dashboards
Customer 360
Technician performance
Automation monitoring
AI oversight for admins
MCP servers compared
Which field service software has an MCP server?
Fieldproxy, Jobber and ServiceM8 ship official MCP servers. The rest rely on community projects, Zapier or a third-party data copy, or have nothing yet. Each row links to the detail and the sources.
| Software | MCP server | AI can change the software | What it is |
|---|---|---|---|
| Fieldproxy | Official MCP | Yes: fields, screens, tables, automations, agents | 145 tools: read, dispatch, comms, documents, photos, monitoring and build. Claude, ChatGPT, Microsoft 365 Copilot and Teams, GitHub Copilot, Cursor, Codex. |
| ServiceTitan | Third-party only | No | No official MCP server. Community-built servers and Zapier's MCP connect through the ServiceTitan API. |
| Jobber | Official MCP | No | Official connector for Claude and ChatGPT, launched September 25, 2026, on the Core, Connect, Grow and Plus plans. |
| ServiceM8 | Official MCP | No | Official hosted MCP server at api.servicem8.com/mcp, for ChatGPT and Claude, with OAuth sign-in. |
| Housecall Pro | Third-party only | No | No official MCP server. Open-source community servers and Zapier's MCP use the Housecall Pro API. |
| Simpro | Third-party only | No | No official MCP server. SyncHub, a third-party data company, offers a simPRO MCP connector. |
| JobNimbus | Third-party only | No | No official MCP server. Community servers and Zapier's MCP use a JobNimbus API key. |
| AccuLynx | Third-party only | No | No official MCP server. A community server brings AccuLynx jobs, contacts and estimates into Claude Desktop and Cursor. |
| Workiz | Third-party only | No | No official MCP server. An open-source server published by a Workiz customer exposes jobs, leads and team data. |
| Service Fusion | Third-party only | No | No official MCP server. Zapier's MCP connects to Service Fusion's Zapier actions. |
| Tradify | None found | No | We found no official or community MCP server for Tradify. |
| FieldPulse | None found | No | We found no official or community MCP server for FieldPulse. |
| BuildOps | None found | No | We found no official or community MCP server for BuildOps. |
See it built around how your team already works.
Get DemoGuardrails
Your AI proposes. A person confirms.
Built for operations with branches, teams and an IT review, not a single login.
Nothing changes on one call
Every write, send or action is a proposal the user confirms within 30 minutes. Writes can be undone. More than 5 rows needs bulk permission, and more than 1,000 is refused.
Keys scoped like API keys
Limit a key to certain tables, certain rows ("only the North region's jobs"), hidden columns, insert/update/delete, and an expiry from 1 hour to 90 days. Postgres enforces the row limits.
Builds save as drafts
Apps and automations are checked against the database before they save, new automations and agents are saved switched off, and public apps are never changed.
Sends are limited
Email goes only from the user's own mailbox, at most 20 sends an hour per key, inside the workspace's quiet hours. A sandbox never sends.
Everything is logged
Every call is written to an audit log with the tool, key, person and outcome. Administrators see every key and connected app, and revoking one applies on the next request.
Access levels
Three levels, and your own role is the ceiling
Each connection gets one or more scopes. A person can never grant an AI more than they can do themselves.
| Scope | Requires | Unlocks |
|---|---|---|
| Read (data:read) | AI assistant access on for the workspace, and the user has Command Center access | Your schema, ready-made reports, and read-only SQL on your own data |
| Write (data:write) | Tables the workspace lets the AI write, and no approval gate on them | Changes to records, confirmed by a person and undoable; email and SMS drafts, sent on confirm |
| Build (build) | An administrator, plus Allow build switched on for the workspace | New fields, tables, screens, automations, voice and text agents, and sandboxes |
A key can be narrowed further, like an API key: an expiry from 1 hour to 90 days, only certain tables, only certain rows (for example one region's jobs, the key owner's team, or a service zone), hidden columns, and any mix of insert, update and delete. Limits can be described in plain words, and Fieldproxy shows how many rows the key would see before it is created.
Deliberately not possible
What the AI can never do, however it is asked
- Change anything on a single call: every write, send or action is a proposal a person confirms within 30 minutes
- Change more than 1,000 records in one go (more than 5 needs bulk permission)
- Drop a table or a column: builds only ever add
- Publish or change a public app
- Read a hidden column, however the query asks for it
- See data the signed-in user cannot see: their own access is the ceiling
- Send email from anyone's mailbox but the user's own, or more than 20 sends an hour per key, or outside quiet hours
- Send anything at all from a sandbox
- Run JavaScript tools or tools tied to an app screen (those stay inside Fieldproxy)
- Call a tool that is not in its list, even by name
Security
How the connection actually works
Every request starts from scratch
Nothing is kept between calls, so a changed setting or a revoked key applies on the very next request.
Sign-in on every call
Each request carries a key (fpmcp.…) or a sign-in token that lasts 1 hour and renews itself for 30 days. The key is looked up every time, so a revoked or expired one is refused at once.
A tool list built per person
The tools the AI sees are assembled for each request from the key's scopes, the person's role, the workspace settings and the key's own limits. A tool missing from the list is refused even when called by name.
Reads run locked down
Reads run as a read-only database role, one statement at a time. Postgres applies the key's row limits and hidden columns are removed from every result.
Changes become proposals
Writes, sends, runs and builds are shown to you first and happen only after you confirm. Builds are saved as drafts, and new automations and agents are saved switched off.
Everything is on the record
Every call is logged with the tool, key, person and outcome. Administrators can read the log, see every key and connected app, and revoke any of them.
Want it connected to a workspace built on your data?
Get DemoHow it works
Three steps to connect
Switch it on
Fieldproxy turns on AI assistant access for your workspace, and optionally Allow build. It is off by default.Connect a client
Add the connector in Claude, ChatGPT or Copilot Studio with the Fieldproxy address from Settings, Connect an AI assistant, or create a scoped key (fpmcp.…) under Settings, AI assistant.Ask in plain English
Start with a built-in workflow: daily_ops_briefing, plan_the_schedule, triage_my_inbox, why_did_this_automation_fail, and more.
By trade
What operations teams ask it
HVAC
Plumbing
Electrical
Facilities maintenance
Roofing
Pest control
Guides
Questions teams ask before they connect
Can I build my own field service software with Claude, ChatGPT or an AI app builder?
Can AI customise my field service software?
Can ChatGPT replace field service management software?
How do I use ChatGPT or Claude to dispatch and schedule technicians?
What is the best AI scheduling assistant for a field service business?
Can ChatGPT do invoicing for a service business?
What is an MCP server, and what does it do for field service software?
What are real examples of an MCP server in use?
How do I use an MCP server with ChatGPT, Claude or Copilot?
See it built around how your team already works.
Get DemoAI that runs the work
Not a chatbot on the side. The way they run the business.
“This is what AI in field service should look like: not a chatbot on the side, but the way you actually run the business.”
Stephanie Choate, Owner, Pronative
Commercial facility maintenance
“Our clients have been able to make decisions quickly based on the AI output and summaries it provides, which has meant more work flowing through.”
Lisa Turner, Director, Total Risk
FAQ
Field service MCP, answered
- Can an AI assistant customise my field service software?
- With Fieldproxy, yes. Its MCP server lets Claude, ChatGPT or Copilot add fields, build screens, create tables, and set up automations and agents. Changes are built as drafts in a sandbox copy of the workspace and go live only after an administrator approves, with rollback. The other field service MCP servers we checked (Jobber, ServiceM8 and community servers) can read and edit records but not change the software itself.
- Can ChatGPT or Claude update hundreds of jobs or contracts at once?
- With Fieldproxy, yes. From one request the assistant can reassign a technician's jobs and text the customers, reprice every maintenance contract, close stale quotes, reschedule a region or import up to 1,000 spreadsheet rows. Each change is dry-run and shown row by row with before and after values, applied only when a person confirms, and can be undone as one change. More than 5 rows needs bulk permission and more than 1,000 is refused.
- What is an MCP server for field service software?
- MCP (Model Context Protocol) is the open standard AI assistants such as Claude, ChatGPT and Microsoft Copilot use to connect to business software. A field service MCP server lets the assistant read your jobs, customers and invoices and take actions like scheduling or sending a message, under the permissions you set.
- Which field service management software has an MCP server?
- Fieldproxy, Jobber and ServiceM8 ship official MCP servers. ServiceTitan, Housecall Pro, Simpro, JobNimbus, AccuLynx, Workiz and Service Fusion have community or third-party servers only. We found none for Tradify, FieldPulse or BuildOps.
- Can ChatGPT or Claude run dispatch for my field service business?
- Only when connected to your field service data. Fieldproxy's MCP server gives them tools to spread jobs across technicians, plan daily routes and apply the schedule, with a person confirming the change before it is saved.
- Can Microsoft Copilot and Teams access field service data?
- Yes, through an MCP tool in Copilot Studio. Fieldproxy's MCP server can be added to a Copilot Studio agent and published to Teams and Microsoft 365 Copilot, with each person signing in to their own Fieldproxy account.
- Is it safe to let an AI change my field service data?
- With Fieldproxy, every write, send or action is a proposal a person confirms, writes can be undone, more than 5 rows needs bulk permission, and every call is logged. Keys can be limited to certain tables, rows and columns, and Postgres enforces the row limits.
- How many tools does Fieldproxy's MCP server have?
- 145 built-in tools, plus the workspace's own saved SQL tools and seven ready-made workflows including a daily operations briefing, schedule planning and inbox triage.
Explore Fieldproxy MCP
Start here
Use cases
More use cases
See your AI run your operation
We connect Claude, ChatGPT or Copilot to a Fieldproxy workspace built on your data and run your real questions on it.