Claude Code is simply my favorite AI tool right now. Period. I’ve been using it daily to create random things I never thought I’d need and I’m not entirely sure how I ever managed without it. I have also trusted him a lot during use cases that have absolutely nothing to do with codingand has helped me tremendously rethink what an AI tool can really do for your productivity.
Having said all that, I know there are thousands of users subscribed to Claude’s plans, but they refuse to give Claude Code a proper chance. Some of the excuses I’ve heard are that “it’s just for developers”, “working in the terminal is terrifying”, “setting it up is too complicated”, or “I set it up and had no idea what to do next”. It’s true, I used to think the same. For a long time I thought about navigating to a folder, writing claudius at the terminal, and asking was the best I could do. That is, until I realized that Claude Code could help me get it set up correctly.
Do you want to stay up to date with the latest in AI? The XDA AI Insider newsletter is published weekly with deep dives, tool recommendations, and practical coverage you won’t find anywhere else on the site. Subscribe by Modifying your newsletter preferences.!
I first asked Claude Code to audit my existing setup.
Let Claude Code judge you first
When you use a GUI-based tool, it’s pretty easy to see which settings you already have enabled and which you’ve disabled. After all, there is a clear checkbox or toggle staring right at you. However, with terminal-based tools like Claude Code, it’s not always obvious what you’ve configured and what you haven’t. While one option is to review the documentation or slash commands and try to rebuild everything yourself, there is a much simpler way: just ask Claude Code to tell you. Here is the exact message I used:
Hey, can you audit my current Claude Code setup and tell me what settings or configurations I’m missing that would really make a difference?
Now, note that I created an empty folder just for this purpose rather than working within an existing project, so some of Claude’s findings weren’t really applicable since I use those functions, but not within that empty project. For example, the missing CLAUDE.md and GitHub repositories weren’t really applicable here since I already use them in my real projects. However, the audit brought to light features that I had no idea existed or simply hadn’t bothered to configure.
For example, Claude-memory is a feature I’ve heard a lot about.and although I have it set to the standard Claude chat interface, I never set it up for Claude Code. He also recommended the /fewer-permission-prompts slash command, which I had no idea existed and which reduces those constant permission pop-ups that slow you down. Another setting that he told me benefits advanced users is custom keybinds, which are basically keyboard shortcuts for bar commands that you use all the time so you don’t have to type them every session.
Now the good thing is that I didn’t have to bother figuring out how to set up these recommendations later. Given Claude Code’s agent capabilities, it can actually do the configuration for you right there, in the same session. I asked it to configure memory and run the /fewer-permission-prompts command. He also pointed out that my global configuration only had two commands in the allowed list and recommended expanding those patterns to reduce friction between projects. So I asked Claude Code to fix that too.
I ask Claude Code to also set up MCP servers
Plug it in, I’m not going to
If you think that all Claude Code can do is generate lines of code and turn them into awesome looking tools, then I need to tell you: you are seriously underestimating it. Claude becomes much more powerful when you start connecting it to the tools you rely on daily, and the easiest way to do this within Claude Code is through MCP servers. If you’re not familiar with MCP servers, they are an open standard that allows AI tools to connect directly to external services like GitHub, Slack, Notion, and basically anything with an API.
Since it is an open standard, you will find a mix of MCP servers created by the companies themselves and others created by the community. That said, as powerful as these servers are and the possibilities you can unlock when you pair them with Claude, they can sometimes be tricky to set up and configure. You may need to edit a few JSON files here and there to ensure the MCP server is working properly. In fact, searching for the right server or finding out if there is an official one for the tool you have in mind can also be annoying.
So instead of doing all that, I simply tell Claude Code what service I want to connect and ask him to handle it. It finds the relevant GitHub repository, installs the server, and configures everything, all without me touching a single configuration file. For example, I asked him to connect the NotebookLM MCP server for me (an unofficial one) and found the GitHub repository and set everything up. When it encountered a problem, it would even troubleshoot and resolve it automatically.
Claude Code also configures Skills for me
Teach the teacher to teach himself.
Skills are one of Claude’s most underrated traits.and I really don’t remember the last time I set one up myself. If you’re not familiar with what Claude Skills are, they’re basically saved instructions that tell the tool exactly how you want certain tasks to be performed each time. Like the MCP servers, there are some skills created by Anthropic itself and many that you can find online at X, GitHub, and other places where the Claude community gathers.
Setting up a custom Claude skill within the Claude interface is fairly simple, but within Claude Code, you need to create a markdown file, add certain fields to it in the correct format, and save it to the correct directory. It’s not difficult, but it’s not something I want to bother with. Similarly, if you want to download a skill created by someone else, you must search for it, download the file, and manually place it in the correct folder. Again, it’s not difficult, but it’s not something I want to bother with.
So, you guessed it: I just ask Claude Code to do it for me. Whether I want to create a custom skill from scratch or install one I found online, I simply describe what I need and Claude Code creates the markdown file, formats it correctly, and places it in the correct directory. The best part is that when I find a skill that someone shared on X or GitHub, all I need to do is paste the link, tell Claude Code to install it, and I’m done in seconds.
Claude Code also helps me write CLAUDE.MD files
Dear diary, this is how I work.
I mentioned earlier that Claude Code recommended setting up a CLAUDE.md file when I audited, and while that didn’t apply to my empty test folder, it got me thinking about how I write them in general. Honestly, I used to just run /init, let Claude generate a basic one, maybe modify a line or two, and call it a day. But I started asking Claude Code to read my codebase and write a suitable CLAUDE.md based on what he found.
It made a bigger difference than I would have thought! He spotted patterns I hadn’t really thought about, and the sessions that followed went significantly smoother because Claude already understood the conventions of my project. Now, whenever I start a new project, the first thing I do is ask Claude Code to write CLAUDE.md for me according to the instructions I’ve shared. And as the project progresses, I ask you to update the file to reflect the changes.
Best way to configure Claude Code? Ask Claude Code
If a tool is good enough to build with, it should be good enough to build itself. After all, if Claude Code is used to create Claude Code, it can definitely take care of setting it up for you.
