Act on failed inspections

Open in Claude Open in ChatGPT llms-full.txt

When a driver fails an item in an inspection, Fleet records the answer with has_issue: true and raises an issue. There are two ways to pick those up.

Scopes: inspections:read, issues:read (and issues:write to update them).

Poll for new issues

The simplest: every few minutes, ask for issues reported since your last check.

curl "https://api.get-fleet.com/v1/issues?reported_from=2026-11-01T06:00:00Z&status=open&expand=vehicle&sort=reported_at" \
  -H "Authorization: Bearer $FLEET_API_KEY"

When your system has dealt with one, update it:

curl -X PATCH https://api.get-fleet.com/v1/issues/$ISSUE_ID \
  -H "Authorization: Bearer $FLEET_API_KEY" -H "Content-Type: application/json" \
  -d '{"status": "resolved", "resolution": "Bulb replaced under job 88213"}'

Read the full inspection

To see every answer, including photos and readings, expand the responses:

curl "https://api.get-fleet.com/v1/inspections?submitted_from=2026-11-01T00:00:00Z&expand=responses,vehicle" \
  -H "Authorization: Bearer $FLEET_API_KEY"

Each response has the item label, the value (for example "fail"), structured data, and has_issue.