Json data analysis

Skill cxcscmu/SkillLearnBench/skills/b1-one-shot-claude-sonnet-4-6/enterprise-information-search/json-data-analysis

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install

npx -y skills add cxcscmu/SkillLearnBench --skill json-data-analysis

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Python patterns for analyzing large JSON datasets to find specific information, track tokens, and write answers in the required format.

SKILL.md

1.6 KB, as published. Nobody here has run it

JSON Data Analysis

Loading and Analyzing Large JSON Files

import json

with open('/path/to/file.json') as f:
    data = json.load(f)

Counting Tokens (Approximate)

import json

def count_tokens(text):
    """Approximate token count: ~4 chars per token"""
    return len(str(text)) // 4

# Or use tiktoken for more accurate counting
# pip install tiktoken
import tiktoken
enc = tiktoken.get_encoding("cl100k_base")
tokens = len(enc.encode(str(data)))

Writing Answer File

import json

answers = {
    "q1": {"answer": ["emp001", "emp002"], "tokens": 150},
    "q2": {"answer": ["emp003"], "tokens": 200},
}

with open('/root/answer.json', 'w') as f:
    json.dump(answers, f, indent=4)

Searching Nested Structures

def deep_search(obj, keyword):
    """Recursively search for keyword in nested structure"""
    results = []
    if isinstance(obj, str):
        if keyword.lower() in obj.lower():
            results.append(obj)
    elif isinstance(obj, dict):
        for v in obj.values():
            results.extend(deep_search(v, keyword))
    elif isinstance(obj, list):
        for item in obj:
            results.extend(deep_search(item, keyword))
    return results

Extracting Employee IDs

import re

def extract_employee_ids(text):
    """Extract employee IDs matching pattern like EMP001, E001, etc."""
    return re.findall(r'\b[A-Z]{1,3}\d{3,6}\b', text)