TDD Red Phase Agent
First stage of a three-agent TDD workflow: writes one focused failing test at a time from GitHub issue requirements, and verifies it fails for the right reason before any implementation exists.
Overview
This is the first of a three-agent TDD workflow (Red, then Green, then Refactor). Its only job is to write a failing test, one at a time, before any production code exists.
How it works
- It extracts the issue number from the branch name, fetches the linked GitHub issue and reads its requirements, comments and linked PRs for edge cases.
- It writes the simplest failing test for a single behavior, never several tests in one pass, and confirms the plan with you before writing anything.
- It runs the test to confirm it fails because the implementation is missing, not because of a syntax error.
- It uses Jest/Vitest for JS/TS, pytest for Python, JUnit 5 with AssertJ for Java/Kotlin, or xUnit/NUnit for .NET, with the Arrange-Act-Assert structure throughout.
- It names the test after the behavior and issue number, so it's clear which requirement it covers.
Examples
"Write the first failing test for issue #142: reject signups with
an invalid email format."
Installation
You need VS Code with the GitHub Copilot extension. Consider installing TDD Green and TDD Refactor alongside it, they're meant to be used together.
- Create the agents folder and download the file into your repo:
mkdir -p .github/agents curl -o .github/agents/tdd-red.agent.md \ https://raw.githubusercontent.com/github/awesome-copilot/main/agents/tdd-red.agent.md - Reload VS Code, or open a new Copilot Chat; custom agents under
.github/agents/are picked up automatically. - Pick this agent from the mode picker in Copilot Chat, point it at a GitHub issue, and confirm it writes exactly one failing test.
Related assets
Accessibility Expert
Reviews code and designs against WCAG 2.1/2.2, covering semantics, keyboard and focus behavior, forms, media and dynamic SPA updates, and ships framework-specific examples plus a CI setup for automated checks.
Address Comments Agent
Works through pull request review comments one at a time: applies the minimal fix for each, pushes back on comments that don't make sense, adds test coverage, and commits with a descriptive message before moving to the next comment.
ADR Generator
Turns a decision discussion into a numbered Architecture Decision Record: gathers the missing context, assigns the next sequential ADR number, and writes a structured markdown file with consequences, alternatives and rejection reasons.