Short answer
Connect ChatGPT, Claude or Microsoft Copilot to your field service software through an MCP server, then ask in plain English: "Mike is out sick, move his jobs to the nearest qualified tech." With Fieldproxy, the assistant uses dispatch tools to spread jobs across technicians, plan daily routes with drive times and propose the new schedule, which is applied only after you confirm. Without a connection, ChatGPT can only reason about a schedule you paste in.
Dispatch technicians with ChatGPT or Claude
- 1
Connect the assistant
Add Fieldproxy's MCP server as a connector in Claude, an app in ChatGPT developer mode, or a tool in Copilot Studio, and sign in with your Fieldproxy account.
- 2
Ask for the plan
Describe the change: a sick technician, an emergency call-out, or tomorrow's routes for a region.
- 3
Review the proposal
The assistant shows which jobs move, to whom, and the drive time. Nothing is saved yet.
- 4
Confirm
Confirm within 30 minutes and the schedule is applied. Writes can be undone.
Prompts that work
Be specific about the constraint you care about. These are the asks dispatch teams use most:
- "Give me today's operations briefing: late jobs, unassigned jobs and anything overdue."
- "Mike is out sick. Move his jobs to whoever is closest and show me the plan."
- "Fit this emergency call into today's board without breaking any SLA."
- "Plan tomorrow's routes for the North region and show the drive time saved."
- "Who is free near the Westfield site this afternoon?"
Why a confirm step matters
A schedule touches technicians and customers. Fieldproxy's MCP server never changes it on a single call: the assistant proposes, you confirm. Changing more than 5 records needs bulk permission and more than 1,000 is refused, so one bad prompt cannot rewrite the week.
Pasting a spreadsheet into ChatGPT instead
It works for a one-off, but the plan is stale the moment a job changes, ChatGPT cannot see drive times or technician locations, and someone still has to re-enter every change. Connected, the assistant reads the live board and writes the result back.
Fieldproxy facts on this page come from the Fieldproxy MCP reference. Competitor details are from each vendor's own documentation as of September 27, 2026; see the comparison table and sources.