A coding-assistant skill that makes AI agents lead with the next action, using numbered steps and no filler or preamble.
ykdojo/claude-code-tips
A collection of 45+ practical tips for using Claude Code, plus a custom status line script and the dx plugin of everyday dev skills.
- Stars
- 10,248
- Forks
- 909
- Open issues
- 4
- Last commit
- 25 days ago
What is it?
This repository is a long, tip-by-tip guide to working effectively with Claude Code, Anthropic's terminal coding agent. Topics range from basic slash commands and voice input to context management, git and GitHub CLI workflows, running parallel sessions with worktrees, and isolating risky tasks in containers.
Beyond the written tips, it ships a sample status line script that shows the model, git branch, uncommitted file count, sync status and token usage bar, along with setup instructions. It also includes the dx plugin, a set of reusable skills such as a /handoff command for passing work between fresh agent sessions.
The guide is aimed at developers who already use Claude Code and want to get more out of it, from newcomers learning the basics to experienced users building their own workflows and orchestration setups.
Key features
- 01Covers 45+ tips from basic slash commands to advanced workflows
- 02Provides a customizable status line script with model, git and token usage info
- 03Explains git worktrees and tmux-based control for parallel or containerized sessions
- 04Includes the dx plugin with reusable skills such as /handoff
- 05Offers voice input and context management techniques
Use cases
- Learning Claude Code from basics to advanced usage
- Setting up a status line to monitor context and git state
- Handing off long tasks to a fresh agent session with a handoff file
- Running multiple Claude Code sessions in parallel with git worktrees
Quick start
- 1
Read the tips in the README, starting with Tip 0 for the status line
- 2
Follow the setup instructions in scripts/README.md to install the status line script
Things to know
- The license is unclear (NOASSERTION), so check the repository before reusing the scripts.
- Several voice and tooling tips are Mac-specific, such as the macOS transcription apps.
Topics
FAQ
- What is claude-code-tips?
- A collection of 45+ practical tips for using Claude Code, plus a custom status line script and the dx plugin of everyday dev skills. This repository is a long, tip-by-tip guide to working effectively with Claude Code, Anthropic's terminal coding agent. Topics range from basic slash commands and voice input to context management, git and GitHub CLI workflows, running parallel sessions with worktrees, and isolating risky tasks in containers.
- Is claude-code-tips free to use? What license does it have?
- claude-code-tips is open source under the Other license. Check the license terms before using it in commercial products.
- Is claude-code-tips still maintained?
- Actively maintained. The last commit to claude-code-tips was on Sep 25, 2026.
- What can I use claude-code-tips for?
- Learning Claude Code from basics to advanced usage Setting up a status line to monitor context and git state Handing off long tasks to a fresh agent session with a handoff file Running multiple Claude Code sessions in parallel with git worktrees
- How do I install claude-code-tips?
- Read the tips in the README, starting with Tip 0 for the status line Follow the setup instructions in scripts/README.md to install the status line script
- Where can I get claude-code-tips?
- The source code and installation instructions are on GitHub: https://github.com/ykdojo/claude-code-tips
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A visual, example-driven guide to Claude Code with copy-paste templates for slash commands, memory, skills, subagents, hooks, and MCP.
Two free online books explaining how agent harnesses for code-writing models, such as Claude Code and Codex, are designed and controlled.
An agentic coding tool from Anthropic that runs in your terminal, reads your codebase, and handles coding tasks and git workflows via natural language.
A CLI and skill layer that installs, configures and health-checks channels so AI agents can read and search Twitter, Reddit, YouTube, Bilibili, XiaoHongShu and more.
CLI proxy that filters and compresses shell command output before it reaches an AI coding agent's context, cutting token use.