A reliability engineer at the MAIN site is triaging an open work order that was raised
against a chiller but has not yet been approved for execution. The work order text
reports a suspected anomaly on one specific process measurement, and before approving
field work the engineer wants that sensor history checked with a time-series anomaly
detector rather than assessed by eye.

Carry out that check end to end:

1. Find the work order at the MAIN site that is still awaiting approval and asks for an
   anomaly to be investigated. Identify the asset it concerns and the exact measurement
   its description names.
2. Look up that asset in the asset registry and use the name recorded there.
3. Retrieve the full available sensor history for that asset, limited to that single
   measurement.
4. Write the retrieved series to a file on disk in a form the time-series tooling can
   read. Do not paste the series into your reasoning or into a tool argument.
5. Run a time-series anomaly-detection recipe over that file and obtain dense anomaly
   labels for the series.
6. Report the outcome.

Count as observations only those records that carry a value for that measurement;
records where it is absent are not part of the analysed series.

Return only a single JSON object with exactly these keys and no other text:

{
  "work_order": "<work order number>",
  "asset": "<asset name as recorded in the asset registry>",
  "sensor": "<exact sensor name>",
  "observations": <number of observations analysed, integer>,
  "start": "<first timestamp, ISO 8601>",
  "end": "<last timestamp, ISO 8601>",
  "anomalies_found": <true or false>
}