🤖 CodeCred AI | AI Skill Verification Tool
CodeCred AI is an AI-powered skill verification tool that compares a candidate’s resume claims with their actual GitHub work.
It analyzes repositories, dependencies, and coding patterns to identify real technical skills and highlight gaps between what someone claims and what their code demonstrates.
This makes it useful for:
- Recruiters validating developer skills
- Students checking portfolio authenticity
- Developers showcasing verified expertise
🚀 Key Features
🔎 Skill Verification from GitHub
- Scans repositories, languages, and dependency files.
- Detects frameworks, databases, and technologies used in projects.
- Matches them against resume skills.
🧠 AI-Powered Insights
- Uses Google Gemini AI for intelligent analysis.
- Generates a structured report including:
- ✅ Verified Skills
- ⚠️ Claimed but Not Found Skills
- 💡 Additional Skills Detected
- ✅ Verified Skills
🛠️ Tech Stack
Backend
- Python (Flask)
AI Engine
- Google Gemini API
Data Processing
- GitHub REST API
- PDF text extraction (
pypdf)
Frontend
- HTML5 + Bootstrap 5
📂 Project Structure
CodeCred-AI/
├── app.py
├── requirements.txt
├── .env
└── services/
├── ai_service.py
├── github_service.py
└── pdf_service.py
⚡ Setup & Installation
1. Clone Repository
git clone https://github.com/yourusername/codecred-ai.git
cd codecred-ai
2. Install Dependencies
pip install -r requirements.txt
3. Configure Environment Variables
Create a .env file:
GITHUB_TOKEN=your_github_token
GEMINI_API_KEY=your_gemini_api_key
4. Run the Application
python app.py
Server runs at:
http://127.0.0.1:5000
📊 How It Works
- Extracts skills from the resume PDF.
- Fetches GitHub repositories and code metadata.
- Detects technologies, frameworks, and coding patterns.
- Uses AI to generate a verification report comparing claims vs reality.
Log in or sign up for Devpost to join the conversation.