Fieldproxy guide

How do I find out why a field service automation failed, using AI?

Updated September 27, 2026

Short answer

Ask the assistant. Fieldproxy's built-in why_did_this_automation_fail workflow lets ChatGPT, Claude or Copilot list failed runs, explain the cause, and then retry or dismiss them, simulate a change, or restore an earlier version of the rule. Retries and restores are proposals you confirm.

What to ask

  • "Why did last night's invoice automation fail?"
  • "Retry the three failed runs."
  • "Would this change have worked?"
  • "Go back to yesterday's version of the rule."

How it works

The workflow uses list_failed_runs and list_automation_runs, then retry_automation_run or dismiss_failed_run, simulate_automation, and list_automation_versions with restore_automation_version.

Anything that changes data or sends a message is shown to you first and happens only after you confirm; writes can be undone, and every call is logged.

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.

Illustration of Fieldproxy's MCP tools inside Claude, ChatGPT and Copilot. Names and data are examples.

FAQ

Questions

Can the AI rewrite the automation?
With the build scope, an administrator can have it change the rule; the change is validated and saved as a draft, switched off until turned on.

See it on your operation

We connect Claude, ChatGPT or Copilot to a Fieldproxy workspace built on your data and run your real questions.