AICodeKing connects the open-source Qwen Audio Agent to Claude Code so he can keep a voice conversation going, interrupt answers and change requirements while a VoiceDesk app is built in the background.
Run your coding agent off GitHub issues instead of chat prompts
Leon van Zyl logs and labels GitHub issues with dependencies for every piece of work, then tells Claude Code to read the outstanding issues and implement them one by one.
Leon van Zyl · YouTube · published Oct 29, 2026 · Software & DevOps
▶ Watch on YouTubePrompt
Read all the outstanding issues on my GitHub repository [repo name] and implement them one by one, respecting the labels and dependencies between them.
Visual breakdown
Leon keeps a GitHub repository for every project and logs each piece of work as a GitHub issue, written by him or by the agent. Claude Code is then told to read the outstanding issues and implement them one by one.
The problem: Work requested in chat is hard to track. You want one place that shows the status of all your work and that the agent can pick up from.
From issue to implemented work
- GitHub repositorySet up for every new or existing project
- Log an issueYou write it yourself, or ask your agent to create it
- Label and linkAdd labels and set dependencies between issues
- Claude CodeReads all outstanding issues on the project
- Implement one by oneWorks through the issues in order
- Finished workStatus of all work is visible in the issue list
Source of truthManual setupAI agentResult
What makes issues useful
- Issues can be labeled
- Dependencies can be set between them
- They show the status of all your work
By the numbers
Try it on your own project
Start with a GitHub repository, then route every request through issues instead of one-off chat prompts.
- Make sure the project has a GitHub repository
- Log a GitHub issue for the next piece of work, or have your agent create it
- Label the issues and set dependencies between them
- Open Claude Code and ask it to read all outstanding issues
- Tell it to implement them one by one
Leon says this builds toward a world where you, or even your users, only log a bug or issue and coding agents run in the background.
Takeaways from the clip
- Do this for new projects and existing ones alike.
- The agent can write the issues for you, so logging them costs almost nothing.
- Labels and dependencies keep the order of work clear.
- The goal is that you don't have to interact with Claude Code yourself for every task.
Jump to the part you need
Distilled from the video’s captions
How to try it
- Open Claude Code.
- Copy the prompt and fill in the [placeholders].
- Watch Leon van Zyl's video for the full walkthrough and setup.
Related use cases
Skill Leap AI gives Claude Code a prompt and the path to about 400 memory and knowledge files, and gets a 3D castle with rooms, a memory wall and a bookshelf of clickable books.
Igor from The AI Advantage points ChatGPT (or Codex/Claude) at the open-source Reverse Engineer Anything repo, inside a project with folder access, and asks it to make a version of an app he paid for. He recreated Screen Brush as AIA Ink and a Slido clone.
Income Stream Surfers shares a free Claude Code prompt that audits a repo, mocks up a cleaner UI for approval, regroups menus into 4-7 areas, and ships it behind a feature flag.