Build a dev-workflow generator with test-driven agent coding
The author built an open-source system that generates project-specific dev workflows from an interview, entirely with Claude Code, which wrote and debugged 550+ tests.
Done withClaude Code
- What they did
- The author says they built the whole system through Claude Code sessions: after a /bootstrap interview about a project's stack, it generates project-specific slash commands, guard hooks, pre-push checks, and agent configs. Claude Code also wrote the main code files, produced and debugged 550+ tests in a test-driven workflow, ran self-audits with four parallel review agents, and, via /task-hunter plus Backlog.md CLI, picked tasks, implemented changes, committed, and closed them autonomously.
- How it went
- The result was an open-source, MIT-licensed workflow system the author says spans 15k+ lines, 200+ tasks, 20+ hook files, and supports many common web stacks plus a generic mode. The stated shortcoming was that Claude Code needed guardrails: without hooks, task structure, and automated review, it wrote to wrong directories, lost context, and let subtle bugs through.
- Worth knowing
- To try it as described, you need both Claude Code CLI and Backlog.md CLI, then point the template at your codebase and run /bootstrap.
Try it yourself with Claude Code
Help me build [tool that does X] in [language] using test-driven development. For each feature, write failing tests first, then the code to make them pass, and run the full suite after each step, debugging any failures yourself. Stop when [the feature list] is implemented and all tests pass, then summarize the test coverage.
Want help getting this working? Join the Vibe Coding Academy ↗ Run by Alex Finn, who also runs this site.
Source: Reddit · Apr 1, 2026
Related use cases
Turn tickets into merged pull requests
Pulls tasks from GitHub Issues or Linear, then runs Claude Code or Codex in isolated Kubernetes pods to carry each one through to a PR.
Design and build a robotic arm with no CAD experience
A builder used Claude Code with FreeCAD and Claude Vision to design, print, and assemble a 5-DOF servo arm with vision and voice control.
Hand off coding sessions between agent accounts at usage limits
A developer built an app that tracks usage across multiple Codex and Claude Code accounts and auto-hands-off conversations when one hits its limit.
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.
Reverse engineer legacy systems without source code
Thoughtworks consultants used Gemini 2.5 Pro in a two-week pilot to produce validated blueprints of legacy systems' functionality.
Audit a feature's issues and hand them to an engineer's agent
An AI agent listed all issues with a feature and sent them to an engineer, whose own agent then addressed most of them automatically.
Custom agent
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.