Claude Code PRO workflow
A Russian-language Claude Code course of 12 modules and 8 labs with a ready kit of skills, subagents and hooks for a project
- 330 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The kit adds hooks with shell scripts that run on file edits and at session start
- The `settings.json` file sets the agent's permissions and `.mcp.json` connects GitHub and Playwright, so read and adapt them for your project
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add workflows/claude-code-proDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The kit adds hooks with shell scripts that run on file edits and at session start
- The `settings.json` file sets the agent's permissions and `.mcp.json` connects GitHub and Playwright, so read and adapt them for your project
This is third-party code. Review the repository files before installing.
What it does
Claude Code PRO teaches Claude Code in order: from installation and first prompts to CLAUDE.md, permissions and hooks, MCP and CI. The course has 12 modules and 8 labs with done criteria, plus a cheat sheet and an FAQ. The theory comes with a starter-kit, a ready configuration to copy into a project. It holds a CLAUDE.md template, path rules for tests, API and frontend, and a `settings.json` file with permissions. It also has eight skills, five subagents and four hook scripts. The skills are called with `/code-review`, `/commit`, `/write-tests`, `/fix-issue`, `/refactor`, `/pr-description`, `/release-notes` and `/onboard`. The subagents are spread across models: `architect` on Opus read-only, `code-reviewer` on Sonnet, `test-runner` on Haiku. The hooks block edits to `.env` files, lock files and migrations, format code after edits and show the branch and recent commits at session start. The kit includes a `.mcp.json` with GitHub and Playwright. The `install.sh` script copies the kit into a project, and with `--user` it puts skills and agents into ~/.claude for all projects. The repository also holds a plugin example, a prompt library for 13 task types and GitHub Actions workflows for @claude mentions in issues and reviews.
Who it is for. For developers who have tried Claude Code and want a proper setup, and for team leads who need shared rules for a team.
Good fit when
- You are starting with Claude Code and want a path from installation to hooks and CI with labs for each module
- A team needs shared rules: CLAUDE.md, permissions and hooks in one kit
- A project needs ready `/commit`, `/code-review` and `/write-tests` commands without writing skills from scratch
- You want a plugin example to package your settings and share them with colleagues
- You need a template for automatic review and @claude replies in issues through GitHub Actions
Not a fit when
- You work in Cursor or Codex: the course and the kit target Claude Code only
- You need an official Anthropic reference: the course is written by a third party and may lag behind new versions of the tool
- You only want to install one skill or MCP server: a separate catalog entry fits better
- The machine has no jq utility: the hooks and the status line do not work without it
Example request
Install the starter-kit from the Claude Code PRO course into my project without overwriting existing files, and explain which hooks it enables.Limitations
The materials are in Russian and structured around Claude Code, and they do not transfer to other agents without edits. The hooks and the status line need jq. With `--force` the install script keeps old file versions as backups. The course comes from a third-party author and may lag behind new Claude Code versions.
The kit contains a `settings.json` with permissions, hooks and a `.mcp.json`, so read them and adapt them before using them in a working project. The GitHub entry in `.mcp.json` takes its token from the `GITHUB_TOKEN` variable. Anthropic does not sell Claude subscriptions in Russia.
How to disable. Delete the copied files from the project: the `.claude` folder, CLAUDE.md and `.mcp.json`. `.bak` backups appear only when you run with the `--force` flag.
FAQ
Where do I start with the Claude Code PRO course?
With module 00 on installation and first launch, followed by prompting basics and CLAUDE.md. The labs are tied to modules: the first to CLAUDE.md, the second to permissions, the third to test-driven development.
How do I install the starter-kit in a project?
Run `bash install.sh /path/to/your-project` from the cloned repository. Existing files are not overwritten. By hand, `cp -r starter-kit/.claude starter-kit/CLAUDE.md starter-kit/.mcp.json your-project/` does the same.
How does installing with the `--user` flag differ from installing into a project?
`bash install.sh --user` puts skills, agents and output styles into ~/.claude, and they work in all projects. Installing into a project copies CLAUDE.md, rules, hooks, settings and `.mcp.json` into that folder only.
Which subagents does the kit include?
Five: `architect` on Opus, `code-reviewer` on Sonnet, `test-runner` on Haiku, `debugger` on the parent session's model and `docs-writer` on Sonnet. The architect and the reviewer are read-only.
What does the `protect-files` hook protect?
The `protect-files.sh` script fires before the Edit and Write tools and blocks edits to `.env` files, lock files and migrations.
Discussion
No questions or reviews yet.
Related
WorkflowsA learning kit for Claude Code: feature walkthroughs plus copy-paste templates for commands, skills, memory, hooks and subagents
Workflows for Claude Code: /do, /omo, /bmad-pilot and more commands, with Codex, Gemini or OpenCode writing the code through codeagent-wrapper
A Claude Code plugin marketplace with ready sets of agents, slash commands, hooks and skills, plus a search site
The author's personal Claude Code toolbox: meta-prompt and todo commands, skills to create skills, hooks and MCP servers, auditors and a Ralph loop
A set of /sc slash commands, agents and behavior modes for Claude Code, from brainstorming and research to implementation and tests
A Claude Code reference plus a collection of over 70 skills for development, security, data and docs, with agent configs
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| Claude Code PRO (this entry) | no | 330 | |
| My Claude Code Setup | no | 2,701 | a CLAUDE.md memory-bank template with ready hooks, subagents and skills, but no teaching modules or labs |
| Skills for real engineers by Matt Pocock | no | 306,265 | small composable skills for engineering with agents, without a course or overall project setup |
README in short
Besides the modules, the course covers the agent loop, working principles and antipatterns and gives tips for saving context. The recommended study route is laid out on the repository's front page.