Use specs to guide Copilot through backend endpoint work
A developer used Copilot to execute backend work (new endpoint, auth logic update, webhook handling, tests) once a clear spec defined the scope.
Done withGitHub Copilot
- What they did
- On a small SaaS backend task, the developer first wrote a one-screen spec listing the goal, non-goals, allowed files, constraints, and acceptance checks. With that scope set, they used Copilot for the implementation work and repetitive edits across things like a new endpoint, auth changes, webhook handling, and tests, while other chat models helped draft the spec and think through edge cases.
- How it went
- They say this reduced drift versus letting Copilot run from a vague TODO and helped keep fast code generation aligned to existing patterns. The stated limitation is that Copilot still improvises if the task is under-specified, especially on larger changes.
- Worth knowing
- Their rule of thumb was not to delegate until they could write acceptance checks; for bigger changes, they sometimes added a planning layer to break the checklist into file-level tasks.
Try it yourself with GitHub Copilot
Here is my spec: [paste spec covering new endpoint, auth change, webhook handling and tests]. Implement it in my repo, staying strictly within the scope of the spec. Add tests for each part and run them. You're done when the tests pass, and give me a summary of every file you changed.
Want help getting this working? Join the Vibe Coding Academy ↗ Run by Alex Finn, who also runs this site.
Source: Reddit · 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.
Assign GitHub issues to Copilot coding agent
A developer assigned a GitHub issue to Copilot and watched it work, but it started from the default branch instead of the intended feature branch.
Run coding agents from one front-end across machines and VPS
A developer uses T3 Code as a front-end for Claude Code, Codex, and pi on personal machines and a VPS, and Copilot at work, for implementing tasks and drafting commit messages and PR summaries.
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.
Run agents that work Jira tickets, review PRs and send updates
Luke runs several Hermes agents that read Jira tickets, write code, review PRs, track costs, summarize docs and send updates.
Visualize your codebase and coding agents in a 3D workspace
A solo dev built a local-first 3D environment where existing coding agents are embodied as avatars, and the repo is shown as a city to browse code and run apps.
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.