Claude Command Suite plugin
A set of slash commands, skills and subagents for Claude Code: review, tests, security, releases, docs and GitHub to Linear sync
- 1,350 ★ on the repository
- Last commit
Broad access
Why this level
- Commands run git, builds, deployments and release rollbacks, and `remove-dead-code` deletes project files
- Skills read Cloudflare and ElevenLabs keys from `.env`, work with Linear and send audio to the ElevenLabs cloud
- The `/dev:watch` command and the `file-watcher` skill run `claude -p` on a file change event without your involvement
Install
- Works in
Manual install
mkdir -p .claude/commandsDirect install: then download the command files you need into this folder or copy them from the cloned repository.
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Commands run git, builds, deployments and release rollbacks, and `remove-dead-code` deletes project files
- Skills read Cloudflare and ElevenLabs keys from `.env`, work with Linear and send audio to the ElevenLabs cloud
- The `/dev:watch` command and the `file-watcher` skill run `claude -p` on a file change event without your involvement
This is third-party code. Review the repository files before installing.
What it does
Claude Command Suite is a ready-made `.claude` folder for Claude Code: Markdown files with commands, subagents and skills that you drop into a project or into `~/.claude`. Each command runs as `/namespace:command`, and the text after it is placed into the template through `$ARGUMENTS`. For example, `/dev:code-review` walks the agent through eight steps: repository analysis, code quality, security, performance, architecture, test coverage, documentation and final recommendations. Commands are sorted into 25 namespaces. They include `dev` for debugging and refactoring, `test` for tests and coverage, `security` for dependency audits, `deploy` for release preparation and rollback, `docs` for documentation, `sync` for exchanging tasks between GitHub and Linear, `orchestration` for splitting a project into tasks, `webmcp` for wiring WebMCP in Chrome and `media` for video frames and audio transcription. There are also `rust` and `svelte` with stack-specific checks and `simulation` with commands for scenario analysis of decisions. There are twelve skills and, by the author's count, 54 agents. Skills switch on by context: `cloudflare-manager`, `linear-todo-sync`, `remove-dead-code`, `file-watcher`, `gsap-animation` and others. Forty-four agents were copied from the wshobson/agents repository, the rest sit separately.
Who it is for. For Claude Code developers who want a ready set of working commands for every project stage and are willing to pick the folders they need.
Good fit when
- You need ready commands for code review, security audit and dependency audit instead of writing them from scratch
- The project lives in Linear and tasks have to move from GitHub Issues and back with the `/sync:*` commands
- You must prepare a release, a changelog or a step-by-step rollback: the `deploy` commands cover that
- You want to build your own skill and take the `/skills:build-skill` command with five templates as a base
- Dead code has to be cleaned out with a backup branch and a build and test check after removal
Not a fit when
- You do not work in Claude Code: the author lists Claude Code 1.0 or newer as the requirement and the docs mention no other agents
- You need one or two commands: installing the whole set of 216 commands is not worth it, copying the files you need is enough
- A clear code license matters to you: the repository has no license file, so usage rights are not defined
- You need a command for a hand-tuned pre-release check: the ready `deploy` templates give a general order of actions, not knowledge of your infrastructure
Example request
Run /dev:code-review for the src folder and show bugs, security issues and performance bottlenecks with prioritiesLimitations
- The author did not specify a license, so rights to use and copy are not defined; ask the author. The MIT badge at the top of the docs is not backed by anything: there is no license file in the root.
- Numbers in the docs disagree: one place says 54 agents, another 55. The agents folder holds files together with copies from the wshobson/agents repository, more than a hundred in total.
- Some skills need outside keys and runtimes: `cloudflare-manager` needs Bun and `CLOUDFLARE_API_KEY`, `elevenlabs-transcribe` needs `uv` and `ELEVENLABS_API_KEY`, and `webmcp` targets Chrome 146 or newer with a flag turned on.
- The `install.sh` script exits with an error if the `claude` command is not on PATH. If a `.claude` folder already exists, it offers to merge the files, make a backup or cancel.
- The `linear-todo-sync` skill and the `sync` commands work only with a Linear account and access to its API.
How to disable. Delete the copied files from `.claude/commands`, `.claude/agents` and `.claude/skills` in the project or in `~/.claude`. If you installed through `install.sh` with a backup, you can restore the previous folder from the directory with the `.backup` suffix.
FAQ
How do I install only the slash commands I need for Claude Code?
The author calls direct installation the recommended path. Create `.claude/commands` in the project and download the needed file with a command like `curl -o .claude/commands/dev/code-review.md` and the file address on `raw.githubusercontent.com`. `install.sh` installs the whole set and offers a project install into `./.claude` or a user install into `~/.claude`.
How do I sync GitHub Issues and Linear tasks with commands?
The `sync` namespace holds twelve commands: moving tasks both ways, bulk import, linking pull requests to tasks, a health check and conflict resolution. Access to Linear is set up separately; see the file of the command you need for the key and the connection method.
What does remove-dead-code do and is it safe?
Several scout agents look for unused exports, orphaned files and extra imports. Before deleting, the command creates a backup branch, and after deleting it checks TypeScript types, the build and the tests. It removes real files, so read the scout report before you confirm.
How do I run Claude Code when files in a project change?
Use the `/dev:watch` command and the `file-watcher` skill of the same purpose, built on Chokidar. It watches files and folders, filters them by glob, applies a debounce pause and runs `claude -p`, inserting the file path through `{{file}}`. This is how you set up automatic review, type checks or test generation.
How do the suite's skills differ from its commands?
You call a command yourself through `/namespace:command`. Claude turns on a skill by context: the phrase "What do I need to work on today?" starts `linear-todo-sync`, and "Find and remove dead code" starts `remove-dead-code`. Skills live in `.claude/skills` and commands in `.claude/commands`.
Discussion
No questions or reviews yet.
Related
PluginsA Claude Code plugin: 39 commands for PRs, review, commits and specs, nine expert roles and hooks, in eight language versions
A set of Claude Code utilities: background skills, expert agents, slash commands and prompt templates for development
A set of 24 slash commands for the Claude Code CLI: commits, review, refactoring, tests and cleanup without repeating the same prompts
A ready-made Claude Code setup: agents, slash commands, skills, safety hooks, rules and MCP connections in one install
Claude Code Skills and Agents
Claude Code Skills & Agents
A Claude Code plugin bundling skills and sub-agents: deep research, image generation, GitHub work, code review and model switching
wshobson/commands
Claude Code Slash Commands
A ready set of slash commands for Claude Code, split into tools for focused tasks and workflows for multi-step orchestration
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| Claude Command Suite (this entry) | no | 1,350 | |
| wshobson/commands | no | 2,641 | tools and workflows folders by wshobson, while this has 25 namespaces, skills and agents in one set |
| claude-code-sub-agents | no | 1,698 | a subagent collection only, with no commands or skills |
| Agentic Plugin Marketplace by wshobson | no | 41,556 | wshobson's plugin marketplace with agents and skills, 44 agents of this suite were copied from there |
README in short
The docs open with a summary of what the suite contains and a list of capabilities: commands, agents, skills, standalone commands, ready chains and scenario simulators. Next come sections on skills with each one's requirements, two installation methods, a table of namespaces and the full command list with short descriptions. The end collects typical command chains, rules for custom commands, and the automatic changelog and release routine. The repository also has separate guides for installation, customization and development.