Over the holidays, I have a lot of time to engage with technical novelties. And CLI coding was one of them. What triggered me trying to use CLI was Z ai’s new model launch - GLM 4.7. I snatched their $3 monthly coding plan and started coding via opencode. Why opencode? - because I already have a claude code subscription and use Claude models in it. And there was no easy way to use 2 claude code instances with Anthropic and GLM models. Hence opencode. As I worked through my Multi-modal AI app using these models, I got good results from GLM 4.7 and decided to make opencode with GLM coding plan a permanent tool in my AI coding arsenal.

During the holidays, I established all-CLI setup on my machine. Lazyvim, Lazygit, kitty terminal, claude code, opencode, avante plugin with minimax M2.1 model. I am very comfortable with this setup now. As I watched the memory and compute usage of this setup (via htop) against vscode and cursor AI, I am convinced that I can easily stretch the battery life of my laptop even farther it already is.

Another benefit of this setup ? - I am reading a lot more of the code. I don’t know why. It could be the novelty effect that could wear off - but I have strong intuition that it might not. And that is a great thing! I am also more plugged in to git via Lazygit. Every other day I faced some problems with the VSCode extensions and never once did I have any problems with Lazygit.

Other than the setup, what did I learn during the holidays that is now a permanent behavior change? I found out about the Claude Code tool called AskUserQuestion tool. This is a game changer to reduce the gap between me and the code the agents write for me. I feel more connected to the codebase having been interviewed by Claude before implementing a feature. I asked opencode creators if they have anything similar. It turns out opencode may just be 1-2 weeks behind claude code.