Skip to content
Get packs

Credit Risk Data Cleaning

Clean raw credit data and screen variables (missing rate, IV, PSI, noise, correlation) before modeling

Source Download skill
Repository
github/awesome-copilot
Stars
40,703
Licence
MIT
Files
4 files

What it does

This skill gives an AI agent a ready-made pipeline for preparing credit risk data before building a loan model. Once installed, the agent can load a dataset, drop unusable months and features that are mostly empty, unstable, weakly predictive or redundant, and export an Excel report documenting each step. Original data is kept, so results can be compared between steps.

Great for

  • Credit risk modelers preparing data for pre-loan scoring models
  • Data analysts who need a documented, repeatable variable screening process
  • Teams that want an Excel audit trail of why each variable was removed

Skip it if

  • General data cleaning unrelated to credit or lending, since the thresholds and steps are specific to credit risk modeling
  • Making lending decisions directly without model validation and human review, because the skill only filters data and does not validate the model
  • Datasets containing personal data sent to external services, since the skill's network use is not documented in the files and should be checked before running on sensitive data

Try asking

  • “Clean my raw loan application data and remove variables with more than 60% missing values”
  • “Screen features by IV and PSI across organizations and export an Excel cleaning report”
  • “Run the Null Importance denoising and high-correlation removal steps on my modeling sample”
Step-by-step guide: how to use Credit Risk Data CleaningInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 4 original files, the MIT licence and an attribution note.

    datanalysis-credit-risk.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip datanalysis-credit-risk.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip datanalysis-credit-risk.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Review the default thresholds (missing rate 0.6, IV 0.1, PSI 0.1, correlation 0.9) and adjust them to your modeling policy before running.
  • Run on a copy of your data first, since the scripts execute locally and read your files; check the example.py paths match your project layout.
  • Read the Excel report sheets step by step to confirm each removal makes sense before modeling.

What's inside

  • references/analysis.py
  • references/func.py
  • scripts/example.py
  • SKILL.md

Redistributed unmodified from github/awesome-copilot under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.