sentinelci 1.0.3

AI-powered security scanning and autonomous remediation platform

SentinelCI

AI-Powered Security Scanning and Autonomous Remediation Platform

SentinelCI is an advanced security scanning platform that combines AI-powered analysis with autonomous remediation capabilities. It detects vulnerabilities, security misconfigurations, and compliance issues across your codebase and CI/CD pipelines, then automatically creates fixes and pull requests.

🚀 Quick Start

Installation

pip install sentinelci

Initial Setup

# Run the interactive setup wizard
sci onboard

# Or set up manually
sci github setup  # Configure GitHub integration

Basic Usage

# Scan current directory
sci scan

# Analyze GitHub repositories
sci github repos

# Run autonomous security agent
sci github repos
# Select repository → "Autonomous Agent (Full Automation)"

✨ Key Features

🔍 Comprehensive Security Scanning

• Secret Detection: Finds hardcoded API keys, tokens, passwords
• Vulnerability Analysis: CVE scanning with NVD integration
• Dependency Scanning: Identifies vulnerable packages and versions
• CI/CD Security: Analyzes GitHub Actions workflows for security issues
• Code Quality: Detects security anti-patterns and misconfigurations

🤖 AI-Powered Analysis

• Intelligent Threat Detection: AI analyzes context and severity
• False Positive Reduction: Smart filtering reduces noise
• Risk Assessment: Automated severity scoring and impact analysis
• Contextual Recommendations: Tailored fix suggestions

🛠️ Autonomous Remediation

• Automatic Issue Creation: Creates GitHub issues for tracking
• Pull Request Generation: Generates PRs with security fixes
• Code Patching: Applies fixes directly to repositories
• Pipeline Fixes: Corrects CI/CD security misconfigurations
• No Cloning Required: Uses GitHub API for remote operations

🏢 Enterprise Features

• Organization Scanning: Scan all repositories in an organization
• Risk Heatmaps: Visual security dashboards
• Compliance Reporting: Generate security reports
• Integration Ready: Works with existing CI/CD pipelines

📋 Requirements

• Python: 3.11 or higher
• AI API Key: Groq (recommended), OpenAI, or Anthropic
• GitHub PAT: For repository analysis and autonomous features (optional)
• NVD API Key: For enhanced CVE scanning (optional)

🔧 Configuration

AI API Setup

SentinelCI supports multiple AI providers:

1. Groq (Recommended - Fast & Free)

   • Get API key: https://console.groq.com/keys
   • Set: sci onboard or export AI_API_KEY=your_key

2. OpenAI

   • Get API key: https://platform.openai.com/api-keys
   • Set: export AI_API_KEY=your_key

3. Anthropic

   • Get API key: https://console.anthropic.com/
   • Set: export AI_API_KEY=your_key

GitHub Integration

# Set up GitHub Personal Access Token
sci github setup

# Required scopes:
# - 'repo' (for private repositories)
# - 'public_repo' (for public repositories)

Environment Variables

# AI Configuration
export AI_API_KEY="your_ai_api_key"
export GROQ_API_KEY="your_groq_key"  # Alternative

# GitHub Configuration  
export GITHUB_PAT="your_github_token"
export GITHUB_TOKEN="your_github_token"  # Alternative

# NVD Configuration (Optional)
export NVD_API_KEY="your_nvd_key"

📖 Usage Examples

Local Scanning

# Basic scan
sci scan

# Scan with specific severity
sci scan --severity high

# Output to JSON
sci scan --output results.json --format json

# Watch mode (continuous scanning)
sci scan --watch

GitHub Repository Analysis

# List and analyze repositories
sci github repos

# Scan specific organization
sci github scan-org your-org-name

# Check authentication
sci github auth

Autonomous Security Agent

The autonomous agent can automatically:

• Detect security vulnerabilities
• Create fixes and patches
• Generate pull requests
• Open tracking issues
• Apply changes without manual intervention

sci github repos
# Select repository
# Choose "Autonomous Agent (Full Automation)"
# Review and approve the execution plan

Pipeline Security Analysis

# Analyze CI/CD pipelines
sci pipeline analyze .github/workflows/

# Auto-fix pipeline issues
sci pipeline fix .github/workflows/ci.yml

🔒 Security Categories

SentinelCI detects and fixes:

Secrets & Credentials

• API keys, tokens, passwords in code
• Hardcoded credentials in configuration files
• Exposed secrets in environment variables

Dependencies & Supply Chain

• Vulnerable package versions
• Outdated dependencies
• Malicious packages
• License compliance issues

CI/CD Pipeline Security

• Excessive workflow permissions
• Unpinned action versions
• Code injection vulnerabilities
• Missing security checks

Code Security

• SQL injection patterns
• XSS vulnerabilities
• Insecure cryptographic practices
• Authentication bypasses

🏗️ Architecture

┌─────────────────┐    ┌──────────────────┐    ┌─────────────────┐
│   Scanner       │    │   AI Analyzer    │    │  Autonomous     │
│   Engine        │───▶│   (Groq/OpenAI)  │───▶│  Agent          │
└─────────────────┘    └──────────────────┘    └─────────────────┘
         │                        │                       │
         ▼                        ▼                       ▼
┌─────────────────┐    ┌──────────────────┐    ┌─────────────────┐
│   Local Files   │    │   Threat Intel   │    │  GitHub API     │
│   Git Repos     │    │   CVE Database   │    │  Issue/PR Gen   │
└─────────────────┘    └──────────────────┘    └─────────────────┘

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details.

Development Setup

# Clone repository
git clone https://github.com/sentinelci/sentinelci.git
cd sentinelci

# Install in development mode
pip install -e ".[dev]"

# Run tests
pytest

# Format code
black sentinelci/
ruff check sentinelci/

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🆘 Support

• Documentation: https://docs.sentinelci.dev
• Issues: https://github.com/sentinelci/sentinelci/issues
• Discussions: https://github.com/sentinelci/sentinelci/discussions

🙏 Acknowledgments

• Built with Typer and Rich
• AI powered by Groq, OpenAI, and Anthropic
• Security data from NVD and GitHub Security Advisories

---

Made with ❤️ by the SentinelCI Team