Update a code base with Copilot chat and agent mode
A commenter used Copilot's chat and agent mode to update their code base and likes the CLI's autopilot mode and todo tool.
Done withGitHub Copilot
- What they did
- In the comments, one user said they update their code base with GitHub Copilot’s VS Code chat and agent mode, then asked how Copilot CLI compares on premium requests and output quality. The poster replied that CLI uses the same premium-request multiplier logic as the IDE and said they mostly run it in autopilot mode, preferring its output; elsewhere in the thread they describe that style of agent as reading the codebase, planning, writing code, running tests, and opening PRs.
- How it went
- The reported result was qualitative, not measured: the poster said CLI quality is superior for them and that they use autopilot most of the time. No numbers on code changes, time saved, or success rate were given.
- Worth knowing
- Premium-request multipliers were said to be the same in the IDE and CLI.
Try it yourself with GitHub Copilot
Use the Copilot CLI in autopilot mode with a todo list to update my code base: [describe the change, e.g. migrate from library X to library Y]. First write out the todo list and show it to me, then work through it item by item, checking things off as you go. Stop when every item is done and [tests/build] pass, and summarize what 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.
Iterate on tasks with Copilot CLI autopilot and fleet
The author used autopilot to iterate until done and /fleet to split work across parallel subagents, noting overhead on small tasks.
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.
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.
Refactor a million-line repo with 1,393 subagents
A Nous Research developer had Hermes dispatch 1,393 subagents over about 19 hours to clean up its own repo, cutting non-test Python source by 34.4% in a merged PR.
Build a mobile app turning photos and voice into reminders
The poster built Loopi, an app that turns photos, text and voice into reminders, notes and expenses, in about 45 days using Replit and AI agents.
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.