Write your repo's chores as plain-English Markdown and let a coding agent run them inside GitHub Actions: issue triage, CI failure analysis and docs updates, read-only by default: AI agent prompt
Issue triage, figuring out why CI failed and keeping docs current eat time on every repository. On Oct 28, 2026 GitHub made GitHub Agentic Workflows generally available on Free, Team and Enterprise: you describe the automation in a natural-language Markdown file, it compiles into standard GitHub Actions YAML, and GitHub Copilot or your own coding agent does the work. Agents run with read-only permissions by default inside a sandboxed container, outputs go through a 'safe outputs' check and a threat detection job scans proposed changes before they apply; `gh aw forecast` estimates AI credit spend. An Accenture manager quoted by GitHub says one lead saved an estimated 29 hours a week. The prompt below drafts your first workflow file to start from.
The prompt
Help me write my first GitHub Agentic Workflow in plain-English Markdown. My repository: [what it is, language, how tests and CI run]. The chore I want automated: [pick one: triage new issues / explain why a CI run failed / update docs when code changes / write missing tests]. Our rules: [labels we use, who owns what, what a good issue looks like, files that must never change]. 1) Write the workflow as a Markdown file: when it runs (the trigger), what the agent should read, the exact steps it should follow, and the only outputs it's allowed to produce (e.g. add labels and one comment, or open a draft pull request), keeping it read-only otherwise. 2) Add instructions for unclear cases: when to ask a human instead of acting. 3) Write 5 test cases (sample issues or failed runs) and what the right result would be for each, so I can check it before turning it on for everyone. 4) Tell me what to check in the compiled Actions YAML and how to use `gh aw forecast` to estimate cost before I enable it. Don't invent commands or settings you aren't sure exist; mark them so I can look them up in the GitHub Agentic Workflows docs.
Fill in the blanks, then copy
This prompt has 1 [bracket] to fill in. Type your details below and the prompt updates as you go. Your answers stay in this browser and are reused on other prompts with the same blanks. Nothing is sent to us.
Example values are fictional, just to show how a filled-in prompt reads. Replace them with your real details.How to use it
- Copy the prompt above.
- Replace every [bracket] with your own details (use the fill-in-the-blanks box above, which remembers your answers for the next prompt).
- Paste it into GitHub Agentic Workflows (gh aw), GitHub Actions, GitHub Copilot and review the answer before you use it.
Works with: GitHub Agentic Workflows (gh aw), GitHub Actions, GitHub Copilot, Claude Code
Real-world example: github.blog/changelog/2026-10-28-github-agentic-workflows-generally-av
Before you use it: what to check
Already built into this prompt
- Don't invent commands or settings you aren't sure exist; mark them so I can look them up in the GitHub Agentic Workflows docs.
Check yourself
- Check every figure, price and total against your own records before anything is sent, posted or booked. A person approves refunds, discounts, payments and price changes.
- Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details.
Using an AI agent? Download this prompt as a SKILL.md agent skill with these checks built in, or get the .zip for Claude.ai. How to install skills.
Want to make this prompt even better? Grade it with the free prompt grader or build your own with the AI prompt generator.
32 copy-paste agent prompts (PDF) plus 5 new prompts every week.
Get it free