Load and parse DDC CWICR construction cost database from multiple formats: Parquet, Excel, CSV, Qdrant snapshots. Foundation for all CWICR operations.
SkillScore100Maintained
Automatic
◆Data Engineer pack
Last verified 8/27/2026
What it changes
Passed 22 of 22 test cases with the skill, and 4 of 22 without it.
Without skill18%
With skill100%
Checks it passed that the model alone missed
Passes language='es' to load_cwicr.
Calls loader.load on 'cwicr_export.json'.
Passes CWICRFormat.PARQUET as the format parameter.
and 21 more in the full report
18 fixed4 passed either way0 still failing0 broke
Pass-rate lift
+82 pp
Tokens
+139%
Agent turns
0%
Response time
-43%
Model gemini-3.6-flashMeasured 8/27/2026v222 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 18 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 18
Prompt
I want to do a quick one-line load of `TR_workitems_costs_resources_DDC_CWICR.parquet` for Spanish language items without manually instantiating `CWICRDataLoader`. Write the Python call using the provided convenience helper function.
What was checked, without and with the skill
Checked
Without skill
With skill
Calls load_cwicr with source path 'TR_workitems_costs_resources_DDC_CWICR.parquet'.
Passed
Passed
Passes language='es' to load_cwicr.
Missed
Passed
case-11 · gemini-3.6-flash · 8/27/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md464 linessingle file
CWICR Data Loader
Business Case
Problem Statement
DDC CWICR database is distributed in multiple formats:
Apache Parquet (optimized for analytics)
Excel workbooks (human-readable)
CSV files (universal exchange)
Qdrant snapshots (vector search)
Applications need unified data access regardless of source format.
Solution
Universal data loader supporting all CWICR formats with automatic schema detection, validation, and pandas DataFrame conversion.
Business Value
Format agnostic - Load from any CWICR distribution
Validated data - Automatic schema validation
Memory efficient - Lazy loading for large datasets
Type-safe - Proper data types preserved
Technical Implementation
Prerequisites
pip install pandas pyarrow openpyxl qdrant-client
Python Implementation
import pandas as pd
import pyarrow.parquet as pq
from pathlib import Path
from typing import Optional, Dict, Any, List, Union
from dataclasses import dataclass, field
from enum import Enum
import json