From the source
GitHub made Agentic Workflows, an open source project from GitHub Next, generally available on github.com for GitHub Free, Team, and Enterprise.
- The tool lets teams automate reasoning-based tasks such as issue triage, CI failure analysis, documentation updates, and test writing by running coding agents inside GitHub Actions.
- Users define automation in natural language Markdown files, which the project compiles into standard Actions YAML, so workflows reuse existing runner groups and policy constraints.
- Workflows can be powered by GitHub Copilot or another coding agent, and using the default Copilot CLI engine requires a Copilot license.
- GitHub says its own rollout produced measurable results: one lead saved an estimated 29 hours a week, team velocity more than doubled on average, AI authored 91% of commits, and production errors fell from about 11,000 a day to under 1,000.
- Safeguards include read-only permissions by default, sandboxed container execution behind the Agent Workflow Firewall, validated safe outputs, and a threat detection job that scans proposed changes.
- Since the public preview, GitHub added ARC and Docker-in-Docker support, made gh aw forecast non-experimental with threat detection credits and monthly ranges, refreshed the model inventory, and added per-workflow model providers, model routing, and micro-VM agent isolation.
