ShipSafe is an multi-agent system that scans your code, detect vulnerabilities, and generate automated fixes for you.
https://www.youtube.com/watch?v=BzTmxkDC3cU&feature=youtu.be
- Scans GitHub repositories for security vulnerabilities using AI-powered code analysis
- Detects hardcoded secrets, API keys, tokens, and credentials
- Generates automated code patches and remediation suggestions
- Provides detailed vulnerability reports with severity levels and fix recommendations
Connect your GitHub account to start scanning. ShipSafe uses a multi-agent system built on LangGraph and LangChain to analyze your code in multiple stages:
1. Repository Analysis: The system fetches your repository and breaks down each source file using AST (Abstract Syntax Tree) parsing, which identifies individual functions and their exact locations.
2. Vulnerability Detection: Each function is analyzed by CodeBERT, a specialized AI model trained to detect security vulnerabilities. The model assigns a confidence score to flag potential issues like SQL injection, XSS, CSRF, and other common attack vectors.
3. Secret Detection: A parallel agent scans for exposed credentials and API keys using pattern matching and entropy analysis to catch hardcoded secrets that shouldn't be in your code.
4. Context & Remediation: AI agents enrich each finding with severity ratings, root-cause explanations, and automatically generate code patches to fix the issues. You get both the vulnerable code and the suggested fix, side-by-side.
The entire pipeline runs automatically, giving you actionable security insights in minutes.
- Frontend: Next.js, TypeScript, Tailwind CSS
- Backend: Python, FastAPI, LangChain, LangGraph
- AI Models: CodeBERT (vulnerability detection), OpenAI
- Authentication: GitHub OAuth
- Node.js 18+ and npm/yarn/pnpm
- Python 3.9+
- GitHub account
- OpenAI API key (for secret classification)
git clone https://github.com/your-org/ShipSafe.git
cd ShipSafeInstall Python dependencies:
cd backend
pip install -r requirements.txtCreate a .env file in the backend/ directory:
OPENAI_API_KEY=your_openai_api_key
GITHUB_TOKEN=your_github_personal_access_tokenStart the backend server:
python main.pyThe API will be available at http://localhost:8000
Install Node dependencies:
cd frontend
npm installCreate a GitHub OAuth App:
- Go to GitHub Developer Settings
- Click "New OAuth App"
- Set Homepage URL to
http://localhost:3000 - Set Callback URL to
http://localhost:3000/api/auth/callback/github - Copy your Client ID and Client Secret
Create a .env.local file in the frontend/ directory:
GITHUB_CLIENT_ID=your_github_client_id
GITHUB_CLIENT_SECRET=your_github_client_secret
NEXTAUTH_SECRET=your_nextauth_secret # Generate with: openssl rand -base64 32
NEXTAUTH_URL=http://localhost:3000
NEXT_PUBLIC_API_URL=http://localhost:8000Start the development server:
npm run devOpen http://localhost:3000 in your browser.
- Sign in with your GitHub account
- Select a repository to scan
- View vulnerability reports and automated fixes
For detailed setup instructions, see frontend/README.md.