How I can fetch only specific PK from dynamodb to opensearch

So basically I’m trying to set up an open search sync pipeline on aws. Right now, I have the config for the pipeline as (moves data from dynamodb to opensearch):

version: "2"
dynamodb-pipeline:
source:
dynamodb:
acknowledgments: true
tables:
  - table_arn: <>
    stream:
      start_position: LATEST
    export:
      s3_bucket: <>
      s3_region: us-east-1
      s3_prefix: <>
aws:
  sts_role_arn:<>
  region: us-east-1
sink:
- opensearch:
  hosts:
    - https://dummy.amazonaws.com
  index: index
  template_content: |
    {"settings":{"analysis":{"analyzer":{"autocomplete_analyzer":{"tokenizer":"standard","type":"custom","filter":["lowercase","autocomplete_filter"]}},"filter":{"autocomplete_filter":{"type":"edge_ngram","min_gram":2,"max_gram":20}}}},"mappings":{"properties":{"name":{"type":"text","analyzer":"autocomplete_analyzer","search_analyzer":"standard"},"summary":{"type":"text","analyzer":"autocomplete_analyzer","search_analyzer":"standard"},"category":{"type":"keyword"}}}}
  index_type: custom
  document_id: ${getMetadata("primary_key")}
  action: ${getMetadata("opensearch_action")}
  document_version: ${getMetadata("document_version")}
  document_version_type: external
  aws:
    sts_role_arn: <role>
    region: us-east-1


So I don’t want everything from dynamodb, instead I want only PK with value GAMESEARCH to be indexed, I mean the entire dynamodb row record for matching PK, for which I need to specify the template for open search mapping.

Right now the above config kind of tries to copies everything from dynamodb, which I don’t want. Any thoughts?