An intelligent AI-powered bot that automatically analyzes CI/CD failures, provides root cause analysis, and takes corrective actions - saving developers hours of debugging time.

When CI/CD pipelines fail, developers waste 15-20 minutes per failure:

• ❌ Manually checking logs (35,000+ characters)
• ❌ Identifying error patterns
• ❌ Determining if it's a flaky test or real bug
• ❌ Deciding whether to retry or investigate
• ❌ Creating tracking issues
• ❌ Notifying the team

CI/CD Guardian automates all of this in 30 seconds.

CI/CD Guardian is a webhook-based bot that:

1. 🔍 Detects workflow failures in real-time
2. 📥 Downloads and parses logs automatically
3. 🧠 Analyzes failures using AI (Mistral LLM)
4. 🔄 Auto-retries flaky tests (up to 2 attempts)
5. 📝 Creates GitHub issues for persistent bugs
6. 💬 Notifies team via Discord with actionable insights
7. 📊 Tracks metrics and performance

• 85% faster failure triage (20 min → 3 min)
• 70% automated retry handling for flaky tests
• 100% coverage of all CI/CD failures
• Real-time team notifications
• Data-driven insights with metrics

Real-time alerts with AI analysis and clickable links to logs

Detailed failure analysis posted directly on commits

Track performance and success rates in real-time

Automated issue creation with AI analysis and duplicate detection

graph TB
    A[⚠️ GitHub Workflow Fails] -->|Webhook HTTPS| B[🌐 CI/CD Guardian Bot<br/>Flask Server Port 5001]
    B -->|Fetch logs| C[📥 Download Logs<br/>ZIP ~35KB]
    C -->|Extract & parse| D[🔍 Parse Errors<br/>Regex patterns]
    D -->|Send context| E[🧠 AI Analysis<br/>Mistral 7B via Ollama<br/>~10s]
    E -->|Categorize| F{🔀 Failure Type?}
    
    F -->|Flaky Test/Timeout| G[🔄 Auto-Retry<br/>Max 2 attempts]
    F -->|Real Bug| H[📝 Create GitHub Issue<br/>Duplicate detection]
    
    G -->|Update status| I[💬 Discord Notification<br/>Real-time alert]
    H -->|Alert team| I
    
    I -->|Track performance| J[📊 Metrics Collector<br/>8+ KPIs]
    
    style A fill:#ffcdd2,stroke:#c62828,stroke-width:2px
    style B fill:#bbdefb,stroke:#1976d2,stroke-width:2px
    style C fill:#c8e6c9,stroke:#388e3c,stroke-width:2px
    style D fill:#fff9c4,stroke:#f57f17,stroke-width:2px
    style E fill:#e1bee7,stroke:#7b1fa2,stroke-width:3px
    style F fill:#ffe082,stroke:#f57c00,stroke-width:2px
    style G fill:#b2ebf2,stroke:#00838f,stroke-width:2px
    style H fill:#f8bbd0,stroke:#c2185b,stroke-width:2px
    style I fill:#c5cae9,stroke:#303f9f,stroke-width:2px
    style J fill:#b2dfdb,stroke:#00695c,stroke-width:2px

Tech Stack:

• Backend: Python 3.13, Flask
• AI: Ollama (Mistral 7B)
• Integration: PyGithub, Discord Webhooks
• Deployment: Docker, docker-compose
• Monitoring: Custom metrics system
• Testing: pytest (26 tests, 88% core coverage)