# Changelog of CBP WRO & Findings Supplier Screening Monitor (`automa-flow/cbp-wro-findings-supplier-monitor`) Actor

- **URL**: https://apify.com/automa-flow/cbp-wro-findings-supplier-monitor/changelog.md
- **Full Actor documentation**: https://apify.com/automa-flow/cbp-wro-findings-supplier-monitor.md

## Changelog

### 2026-09-29: The primary example is in the README

#### Documentation

- The public example is now in the README, with the input to copy and the Dataset view to open. Replace the sample values with your own. Saved inputs, prices and output fields are unchanged.

### 2026-09-27: Invalid monitor IDs are refused before the run starts (0.1.8)

#### Improvements

- A `monitorId` that is empty or contains characters other than letters,
  digits, dots, underscores and hyphens is now rejected by Apify's input
  validation when you start the run from the Console or the API, so no run is
  created. Previously the run started and then failed with "Invalid global
  input". Valid monitor IDs, and saved Tasks that use them, are unaffected.

### 2026-09-27: Clear failure when the CBP website stops responding (0.1.7)

#### Fixes

- If the CBP website stops responding, the run now stops waiting before its own
  timeout and still returns one row per watch item with `status: FAILED`,
  `match_status: UNKNOWN` and the retryable error code
  `SOURCE_DEADLINE_EXCEEDED`, plus the run summary. Previously such a run could
  reach the run timeout and end with no rows at all. These rows are never
  charged, and monitor mode keeps your last successful check as the baseline.

### 2026-09-16: Unrecognised input fields no longer fail the run (0.1.4)

#### Fixes

- Top-level input fields the Actor does not recognise, for example from an
  older saved Task or an integration, are now ignored with a warning in the run
  log instead of failing the whole run.

### 2026-09-04: First release

#### Features

- Screen up to 5,000 supplier, country, industry and merchandise watch items
  against one validated snapshot of the official CBP Withhold Release Orders
  and Findings dataset.
- Exact entity-name matching that includes CBP's own a/k/a aliases, with
  context screening by country, industry and merchandise keywords.
- Monitor mode compares each watch item with its last successful check and
  reports new, modified and removed actions.
- A source failure is always reported as `FAILED`, never as a non-match, and a
  non-match is never presented as a compliance clearance.