How I use Claude Code? How I bring the best out of Claude Code? What is your workflow?
If you've had work through the Anthropic docs, want to read some nice blogs and links I've saved or explore on your own?
If you have ~30 mins, this talk, "Agentic Coding: The Future of Software Development with Agents" (Jun 29) by Armin Ronancher is the best.
I currently don't sleep a lot
Claude Code is catnip for programmers.
😮
Slides:
What is agentic coding?
Why is this a thing all the sudden?
Cursor vs Claude Code
"Making the $200 Worth It"
Official article: Claude Code: Best practices for agentic coding by Anthropic (Apr 18)
Essential video:
Essential reading:
- https://spiess.dev/blog/how-i-use-claude-code - read this first
- https://lucumr.pocoo.org/2025/6/12/agentic-coding/
- https://harper.blog/2025/05/08/basic-claude-code/
- https://harper.blog/2025/02/16/my-llm-codegen-workflow-atm/
- https://tokenbender.github.io/kautuhal/post.html?id=how-i-bring-the-best-out-of-claude-code
- https://tokenbender.github.io/kautuhal/post.html?id=how-i-bring-the-best-out-of-claude-code-part-2
- https://sankalp.bearblog.dev/my-claude-code-experience-after-2-weeks-of-usage/ (17 Jul)
Optional reading:
- How Anthropic teams use Claude Code (PDF)
- https://www.indragie.com/blog/i-shipped-a-macos-app-built-entirely-by-claude-code
- Poking Around Claude Code
- Claude Code saved us 97% of the work — then failed utterly - Experimenting with Claude Code in CodeConcise, a tool Thoughtworks built to comprehend legacy codebases.
- Yes, Claude Code can decompile itself. Here's the source code. by ghuntley (1 Mar)
- I threw Claude Code at an existing codebase a few months back and quickly quit
- How I use Claude Code (Jun 22)
- How I use Claude Code (25 Jun)
- Mastering Claude Code: Boris Cherny's Guide & Cheatsheet
- How I use Claude Code to implement new features in an existing complex codebase (Jul 6)
- Command your Claude Code Army, Reloaded (2 Jul)
A living document rounding up of some of the blog posts (above): https://steipete.me/posts/2025/essential-reading
Optional video:
- A conversation on Claude Code by Anthropic (Jun 4)
- Claude Code: Anthropic's CLI Agent (YouTube) - podcast
-
Claude Code & the evolution of agentic coding by Boris Cherny, Anthropic (Jul 5)
- fit the workflow to the task
i. explore -> plan -> confirm -> code -> commit
Prompt: figure out the root cause for issue #983, then propose a few fixes. Let me choose an approach before you code. ultrathink
ii. tests -> commit -> code -> iterate -> commit
Prompt: write tests for @utils/markdown.ts to make sure links render properly (note the tests won't pass yet, since links aren't yet implemented). then commit. then update the code to make the tests pass.
This is TDD. It's pretty hard to use in practice. But now with coding tools, it actually works really well. May be the reason is because it's not me doing it, the model doing it.
iii. code -> screenshot -> iterate
Prompt: implement [mock.png]. Then screenshot it with Puppeteer and iterate till it looks like the mock.
If Claude has a target to iterate against, it can do much better. So if there's some way to verify the output like a unit test, integration test, a way to screenshot in your iOS simulator, a way to screenshot in Puppeteer, just some way to see its output.
-
Learn about common workflows with Claude Code in Anthropic docs.
Pricing as of Jul 5, 2025.
- sst/OpenCode - AI coding agent, built for the terminal. The CLI is client-server model (the TUI is their initial focus but the goal is to build alternative frontends, mobile, web, desktop). From their documentations, they recommend signing up for Claude Pro or Max, running OpenCode auth login and selecting Anthropic. It's the most cost-effective way to use OpenCode.
- Great to see more open source agentic CLIs tools.
- Other than the focus on TUI design, does this have any advantage over Claude Code using the same model? OpenCode focused on user experience and less so on performance. OpenCode use all the same system prompts/config as Claude Code. That said people do observe better performance because of out of the box LSP support (edit tools return errors and the model immediately fixes them.)
- If you want to save cost, it's model agnostic. You can switch to another AI provider. You can even use a local Ollama model.
- ripgrim/anon-kode - a fork of the original Claude Code that has been taken down due to copyright issue.
Tools: