GitHub Agentic Workflows

✨ AI-Powered Repository Automation

GitHub Agentic Workflows combines the power of AI agents with GitHub’s collaboration platform to enable Continuous AI — the systematic, automated application of AI to software collaboration.

Run natural language workflows that can understand context, make decisions, and take meaningful actions across your repositories.

Quick Start

Get your first agentic workflow running in minutes. Learn how to install the extension and set up AI-powered automation. Get Started →

Learn Concepts

Understand how agentic workflows work and explore the capabilities of AI-driven repository automation. Learn More →

Browse Commands

Explore all available CLI commands for managing agentic workflows with the GitHub CLI extension. View Commands →

Security Guide

Learn about security considerations, permissions, and best practices for safe agentic automation. Security Guide →

🔧 How It Works

GitHub Agentic Workflows transforms natural language markdown files into GitHub Actions that are executed by AI agents. Here’s a simple example:

---
on:
  issues:
    types: [opened]
permissions: read-all
safe-outputs:
  add-issue-comment:
---
# Issue Clarifier

Analyze the current issue and ask for additional details if the issue is unclear.

The gh aw cli compiles this into a GitHub Actions Workflow (.yml) that runs an AI agent (Claude, Codex, …) in a containerized environment whenever a new issue is opened in the repository.

The AI agent reads your repository context, understands the issue content, and takes appropriate actions - all defined in natural language rather than complex code.

Security Benefits: Workflows use read-only permissions by default, with write operations only allowed through sanitized safe-outputs. Access can be gated to team members only, ensuring AI agents operate within controlled boundaries.

🔥 Key Features

• Natural Language Workflows: Write automation in markdown instead of complex YAML
• AI-Powered Decision Making: Workflows that understand context and adapt to situations
• GitHub Integration: Deep integration with issues, PRs, discussions, and repository management
• Security First: Sandboxed execution with minimal permissions and safe output processing
• Multiple AI Engines: Support for Claude, Codex, and custom AI processors

🎯 Application Areas

Explore application areas of agentic workflows. Each category addresses different aspects of software development and team coordination through intelligent automation.

Research, Status & Planning

Automated workflows for research, team coordination, and project planning. Strategic awareness and team coordination through automated intelligence gathering and progress tracking. Explore Research & Planning Workflows →

Triage & Analysis

Intelligent automation for issue management, accessibility reviews, and CI failure investigation. Deep analytical capabilities with pattern recognition and systematic analysis. Explore Triage & Analysis Workflows →

Coding & Development

Automated workflows for dependency management, documentation updates, and pull request assistance. Streamline development tasks while improving code quality. Explore Coding & Development Workflows →

Quality & Testing

Automated workflows for test coverage improvement and performance optimization. Systematic quality improvements and technical debt reduction. Explore Quality & Testing Workflows →

All sample workflows on this site are part of the Agentics collection - a comprehensive set of demonstrator agentic workflows that showcase the capabilities of GitHub Agentic Workflows for various automation scenarios.

GitHub Agentic Workflows is a research demonstrator from GitHub Next and Microsoft Research. It is in early development and may change significantly. Using agentic workflows requires careful attention to security considerations and human supervision. Use with caution and at your own risk. These are demonstrator workflows for experimental use. Review all outputs carefully and use time-limited trials to evaluate effectiveness for your team.