Write unit tests and add AI review to a CI/CD pipeline
Author used Copilot agent mode to write unit tests and wired Copilot review and failure analysis into CI/CD, reporting CI test failures fell from 18% to 9%.
Done withGitHub Copilot
- What they did
- On a Node.js microservice, the author assigned GitHub issues to Copilot's coding agent. It opened draft PRs with unit tests, and the author left review comments that it then addressed. In week 3 they edited the GitHub Actions workflow to add Copilot review, test-failure analysis and vulnerability-fix steps. Branch protection kept human review in place.
- How it went
- Over two weeks, CI test failures fell from 18% to 9% and average time to merge from 4.2 to 2.1 hours. Failures included a payment refactor that broke 47 tests and a dependency update that got stuck.
- Worth knowing
- The workflow's gh copilot review, fix and analyze-failures commands may not exist as shown, so check them first. The results are the author's own rough figures from two weeks.
Try it yourself with GitHub Copilot
In [owner/repo], write unit tests for [module or folder] to cover the main paths and edge cases, using [test framework]. Then add a CI step that runs Copilot code review on each pull request and summarizes any test failures. You are done when the tests pass locally and in CI, and show me the workflow file changes before you commit them.
Want help getting this working? Join the Vibe Coding Academy ↗ Run by Alex Finn, who also runs this site.
Source: dev.to · Undated
Related use cases
Run a solo company with eight agent departments
A founder set up CEO, CFO, COO, lawyer, accountant, marketing, CTO, and improver agents that share memory and consult each other.
Let Copilot Agent Mode handle coding, then constrain it
A developer let Copilot Agent Mode do all the coding, found they understood their code less, and moved to using it more narrowly.
Edit a large notebook with Copilot Agent Mode
The author asked Copilot Agent Mode for moderate changes and basic features in a large ipynb; it got halfway, stopped mid-line, and broke the notebook.
Build a C compiler with a team of parallel agents
Anthropic's engineering write-up on tasking Opus 4.6 agent teams with building a C compiler.
Ship more pull requests with Devin across tools
Ryan Carson uses Devin from iPhone, Slack, browser, GitHub and Linear for real work and reports higher PR shipping velocity.
Make GitHub profile and repos English-first
Codex inventoried the account, backed up content, and updated the profile, descriptions and README navigation across 87 public repos, then verified the changes against GitHub.
Submitted by SvenKunkka
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.