GitHub Agentic Workflows Examples
GitHub Agentic Workflows (gh-aw) examples show how Markdown workflows can run AI agents through GitHub Actions for repository tasks that require reasoning, interpretation, investigation, or generation. Use this catalog to choose a starting point; each entry explains when the pattern is useful and links to maintained guidance or workflow source.
Examples by task
Section titled “Examples by task”| Task | When to use it | Example |
|---|---|---|
| Issue triage | Classify new issues, identify duplicates, apply bounded labels, and ask for missing information. | AI issue triage on GitHub |
| Pull-request review | Inspect diffs for concrete defects and post review feedback through controlled safe outputs. | Automated AI pull-request review |
| Documentation maintenance | Detect drift between code and documentation and propose reviewable updates. | Keeping documentation up to date automatically |
| CI failure investigation | Analyze failed GitHub Actions runs, correlate logs, and open diagnostic issues with likely causes. | CI Doctor and fault-investigation workflows |
| Code improvement | Find unnecessary complexity or duplicated logic and propose focused changes for human review. | Continuous simplicity workflows |
| Dependency analysis | Research dependency usage and upstream changes before creating prioritized follow-up work. | ResearchPlanAssignOps dependency analysis |
| Repository reporting | Summarize repository or release activity on an event or schedule. | AI-generated release notes and reports |
| Scheduled maintenance | Review a backlog regularly, select bounded maintenance tasks, and propose controlled changes. | Automated repository maintenance |
| Security review | Combine deterministic security tools with AI interpretation to report suspicious changes or compliance work. | Security-related workflows |
Use an example safely
Section titled “Use an example safely”Before enabling an example, review its trigger, AI engine authentication, tools, network access, permissions, and safe outputs. Compile the Markdown source with gh aw compile, inspect both the .md and generated .lock.yml files, and begin with the narrowest permissions and outputs that satisfy the task.
Follow the quickstart to install gh-aw, read Creating Agentic Workflows to adapt an example, compare AI engines, and review the security architecture and FAQ before deployment.