This organization contains:
- CAD application - plugins, addons and tools
- Demonstration projects and starter code projects
- Projects demonstrating Git workflows
- Ideas and experiments
- Software tools and libraries
- Repositories where Claude Coding Agent\Agent SDK performs all software programming activities.
- It means that I will instruct Claude Coding Agent\Agent SDK to perform work, tasks and jobs.
- I will manage, review and test its output.
- My instructions include architecture, solution planning and business rules it must follow.
- Any code I don't understand will still be reviewed and tested through multiple cycles until I am confident in its functionality.
The key difference is that an agent is being used, it performs work autonomously. The agent is being developed for this Github organization and repositories.
This doesn't mean AI/LLM/Agent does all the work, I don't write code or won't take over when there are problems and mistakes it makes. AI/LLM/Agent systems are not perfect and there are many kinds of tasks, problems and technical situations that AI/LLM/Agents are not able to complete.
There are also many tasks that I could complete in minutes that an AI/LLM/Agent could complete but it isn't worth spending tokens, inference or time on instruction/prompt planning for it to do the work.
- Autonomous
- Code reviews, clean up and fixes
- Codebase research and analysis
- Recurring work
There are tasks and work in general that AI/LLM/Agent are not able to complete or solve in a timely manner or at all.
- I manage all repositories under this organization.
- I enjoy building tools and problem solving, I won't stop coding.
- I'll work on what I want and AI/LLM/Agent will do the rest.
These are personal projects and collaborations that you are free to try out.
Stephen
