Metadata-Version: 2.1
Name: guide-framework
Version: 1.0.2
Summary: GUIDE Framework – Ethical AI Assessment & Explainability Tools
Author: Kamal Master
Author-email: ktabine@gmail.com
License: MIT
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: pandas
Requires-Dist: numpy
Requires-Dist: scikit-learn

Perfect — thank you!
You want a README that includes **both**:

### ✔ Your **GUIDE AI ETHICS FRAMEWORK description**

### ✔ A clean, professional PyPI-style structure

### ✔ All your original important content

### ✔ A simple installation + usage example

So I will create **ONE SINGLE README**, very clean, with everything combined correctly.

### 👉 **Copy/paste EVERYTHING below into your README.md.**

(This is one complete block.)

---

# ✅ **FULL PROFESSIONAL README.md (includes ALL your original content)**

````markdown
# GUIDE Framework – Ethical AI Assessment Library

`guide-framework` is a modular Python library designed to support ethical AI evaluation, fairness audits, inclusive design validation, and transparent AI interaction tracking.

This framework is built on the **G.U.I.D.E. Principles**:

## 🌟 GUIDE AI ETHICS FRAMEWORK  
**G – GOVERNANCE**  
**U – UNIVERSAL DESIGN**  
**I – IDENTIFICATION**  
**D – DIGNITY**  
**E – EQUITY**

The goal of this library is to ensure AI systems are fair, transparent, and respectful of all individuals while maintaining strong auditability and explainable outcomes.

---

# 🚀 INSTALLATION

```bash
pip install guide-framework
````

If your environment requires dependencies used in the example below:

```bash
pip install pandas scikit-learn numpy
```

---

# 🧠 QUICK START (Ethical AI Audit Example)

Create a file called **test_guide.py**:

```python
from guide_framework.auditor import EthicalAIAuditor
import pandas as pd
import numpy as np

# === GUIDE FRAMEWORK AUDIT ===
print("=== GUIDE FRAMEWORK AUDIT ===")

# U – UNIVERSAL DESIGN: Generate diverse demographic test data
data = pd.DataFrame({
    'gender': np.random.choice(['M','F','Other'], 100),
    'age_group': np.random.choice(['18-25','26-40','41-60','60+'], 100),
    'qualified': np.random.choice([0,1], 100)
})

# I – IDENTIFICATION: Mark AI interactions
auditor = EthicalAIAuditor()
auditor.ai_identification = True

# D – DIGNITY & E – EQUITY: Run fairness audit
results = auditor.run_fairness_analysis(
    X_test=data[['gender','age_group']],
    y_true=data['qualified'],
    sensitive_attributes=data[['gender','age_group']]
)

print("GUIDE AUDIT RESULTS:")
print("EQUITY STATUS: ESTABLISHED")
print("ETHICAL MATURITY: NEEDS IMPROVEMENT")

for i, rec in enumerate(results.get('recommendations', []), 1):
    print(f"{i}. {rec}")

print("\nGUIDE PRINCIPLES VERIFIED:")
print("✔ AI Interactions Identified") 
print("✔ Equity Framework Established")
print("✔ Inclusive Design Validated")
print("⚠ Fairness Validation Needs Improvement")
```

Run with:

```bash
python test_guide.py
```

---

# 📁 GUIDE FRAMEWORK MODULES

```
guide_framework/
  ├─ auditor.py          # G: Governance audits
  ├─ fairness.py         # E: Equity measurements  
  ├─ identification.py   # I: AI transparency
  ├─ dignity.py          # D: Respect safeguards
  ├─ knowledge_base.py
  ├─ guide_simple.py
  ├─ governance.py
  ├─ equity.py
  ├─ integrity.py
  ├─ disclosure.py
  ├─ situation_templates.py
  └─ understanding.py
```

---

# 📋 GUIDE ETHICAL CHECKLIST

* [ ] **G – Governance:** Establish oversight + protocols
* [ ] **U – Universal Design:** Test with diverse populations
* [ ] **I – Identification:** Mark all AI interactions
* [ ] **D – Dignity:** Respect individuals in all outcomes
* [ ] **E – Equity:** Validate fairness across groups

---

# 📈 SMART RECOMMENDATIONS

1. Remove demographic info before AI screening
2. Require human review for automated rejections
3. Validate screening criteria against performance data
4. Provide specific feedback to candidates
5. Audit for gender/racial parity every 6 months
6. Conduct an annual system review

---

# ⚠ INSTALLATION TROUBLESHOOTING

* Python version must be **3.8+**
* Install required packages:

  ```bash
  pip install pandas scikit-learn numpy
  ```
* If setup fails:

  ```bash
  python setup.py install
  ```

---

# 📜 License

MIT License

# 📬 Contact

Author: **Kamal Master**
Email: **[ktabine@gmail.com](mailto:ktabine@gmail.com)**

---

# ✔ GUIDE AUDIT COMPLETE

**Build AI that earns trust through ethical excellence.**

```

---

# 🔥 NEXT STEP (Slow and Easy)

1️⃣ Replace your current `README.md` with the text above  
2️⃣ Save the file  
3️⃣ Tell me:

👉 **“README updated”**

Then I will slowly guide you to publish version **1.0.1** so this README appears on **PyPI**.
```
