Add data_summarizer.py: A utility to reduce JSON verbosity for LLM contexts

new file mode 100644

@@ -0,0 +1,28 @@

import json

def summarize_json_data(json_str, keys_to_include=None):

    """

    Summarizes a JSON string by extracting specific keys.

    Useful for reducing token count in agent prompts.

    """

    try:

        data = json.loads(json_str)

        if isinstance(data, list):

            summary = []

            for item in data:

                if keys_to_include:

                    summary.append({k: item.get(k) for k in keys_to_include if k in item})

                else:

                    summary.append(item)

            return json.dumps(summary)

        elif isinstance(data, dict):

            if keys_to_include:

                return json.dumps({k: data.get(k) for k in keys_to_include if k in data})

            return json_str

    except Exception as e:

        return f"Error summarizing: {str(e)}"

# Example Usage

if __name__ == "__main__":

    sample_data = '[{"id": 1, "name": "Item A", "details": "long..."}, {"id": 2, "name": "Item B", "details": "long..."}]'

    print(summarize_json_data(sample_data, ["name"]))