Extract a field from a text field

[RaonyO (Raony Oliveira)]

I'm trying to extract the fields that come within the "indicator" and "agentDetectionInfo" field, could someone help me?
here's the log in json:

{
  "_index": "xxx",
  "_type": "_doc",
  "_id": "xxxxxhvfUu_M2V8K",
  "_score": 1,
  "fields": {
    "indicators": [
      "[{'category': 'Abnormalities', 'description': 'This binary contains abnormal section names which could be an indication that it was created with non-standard development tools', 'ids': [1], 'tactics': []}, {'category': 'Hiding/Stealthiness', 'description': 'The majority of sections in this PE have high entropy, a sign of obfuscation or packing', 'ids': [29], 'tactics': []}]"
    ],
    "@timestamp": [
      "2022-05-18T12:23:45.251Z"
    ],
    "agentDetectionInfo": [
      "{'accountId': 'xxxx36xx4xxx4xx3', 'accountName': 'XXX Software', 'agentDetectionState': 'full_mode'}"
    ],
}

[Tomo_M (Tomohiro Mitani)]

Hi,

You may need JSON processor of ingest pipeline to parse the string while indexing, or using Logstash or some ETL client to parse the string before indexing the documents.