myclaude plugin
Workflows for Claude Code: /do, /omo, /bmad-pilot and more commands, with Codex, Gemini or OpenCode writing the code through codeagent-wrapper
- 2,747 ★ on the repository
- Last commit
- AGPL-3.0
Broad access
Why this level
- The installer writes to ~/.claude: CLAUDE.md, commands, agents, skills, a binary and hooks in settings.json, and an update overwrites module files
- The Codex, Gemini and OpenCode backends edit code and run tests on your machine
- The tip to set the Codex approval policy to never turns off its confirmation prompts
Install
- Works in
Manual install
npx github:stellarlinkco/myclaudeAn interactive installer. Options: --list, --update, --install-dir ~/.claude, --force. Files go to ~/.claude, including CLAUDE.md and settings.json.
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- The installer writes to ~/.claude: CLAUDE.md, commands, agents, skills, a binary and hooks in settings.json, and an update overwrites module files
- The Codex, Gemini and OpenCode backends edit code and run tests on your machine
- The tip to set the Codex approval policy to never turns off its confirmation prompts
This is third-party code. Review the repository files before installing.
What it does
myclaude installs a set of ready development processes into Claude Code and distributes roles. Claude Code stays the orchestrator: it plans, gathers context and verifies results. Editing code and running tests is delegated to the codeagent-wrapper utility, which calls one of the backends: Codex, Claude, Gemini or OpenCode. The main `do` module drives feature development through five phases: understand, clarify, design, implement and complete. The clarify phase is mandatory and implementation starts only after your approval. The `omo` module routes a task between specialized agents for code analysis and bug hunting, `bmad` brings an agile process with product owner, architect, developer and tester roles, and `requirements` assembles a quick pipeline from requirements to code. The `essentials` module adds 11 commands: ask, bugfix, code, debug, docs, enhance-prompt, optimize, refactor, review, test and think. Skills for browser automation, test case generation, product requirements, prototype prompts and safe skill installation from GitHub install separately.
Who it is for. For developers who work in Claude Code and want to split planning and coding between different agents using ready workflows.
Good fit when
- A new feature should move through stages with mandatory clarifications and approval before implementation
- A bug is complex and needs an analysis-and-fix route through several specialized agents
- A large project requires product owner, architect, developer and tester roles
- You need a quick prototype from requirements with the /requirements-pilot command
- Part of the work should go to Codex or Gemini, leaving planning and verification to Claude Code
Not a fit when
- You need one simple fix: the /code or /debug commands are enough and a full process is overkill
- The machine lacks the needed CLI agents: the Codex, Claude, Gemini and OpenCode backends each need their own install
- You are not ready for AGPL-3.0 terms: for commercial use without them the author asks you to write to support
- You want a clean setup without hooks: the claudekit module adds hooks to settings.json
Example request
/do add order export to CSVLimitations
The license is AGPL-3.0, and for commercial use without its obligations the author offers contact at support@stellarlink.co. Each backend needs its own CLI agent with certain features: for Codex that is `codex e`, the --json mode and resume, for Claude stream-json, for Gemini the -o stream-json mode.
Installation writes files to ~/.claude, including CLAUDE.md, commands, agents, skills and the codeagent-wrapper binary. The marketplace version is 5.6.1, while the README badge names the 6.x series. The wrapper's source is written in Go.
How to disable. Modules are disabled in config.json with enabled: false. Installed files are deleted from ~/.claude by hand following the list in installed_modules.json, and hooks are removed from settings.json. Claude Code plugins are removed through /plugin.
FAQ
How do I install myclaude in Claude Code?
With npx github:stellarlinkco/myclaude: the interactive installer offers modules and writes files to ~/.claude. --list shows what can be installed, --update refreshes installed modules, and --install-dir with --force change the directory and allow overwriting. Another path is plugins: /plugin marketplace add stellarlinkco/myclaude.
Which module should I pick for a new feature?
For feature development /do is recommended by default, /omo for bug investigation and fixes, /bmad-pilot for a large enterprise project, /requirements-pilot for a quick prototype, and /code or /debug cover simple tasks. This is the selection table from the project description.
What is codeagent-wrapper and why is it needed?
A Go utility through which Claude Code hands execution to another agent. It launches Codex, Claude, Gemini or OpenCode, supports parallel runs and installs to ~/.claude/bin. Claude Code stays the orchestrator while the wrapper executes edits and tests.
How do I enable only the modules I need?
In config.json each module has an enabled flag. The example in the project description enables only do and disables bmad, requirements, essentials, omo, sparv and course.
What if Codex denies permissions?
The project description suggests setting approval_policy to never in ~/.codex/config.yaml. That turns off Codex confirmation prompts, so apply it deliberately and only in an isolated environment.
Discussion
No questions or reviews yet.
Related
PluginsMulti-agent orchestration for Claude Code: teams, roles, autopilot and task routing across skills and subagents
A set of /sc slash commands, agents and behavior modes for Claude Code, from brainstorming and research to implementation and tests
Anthropic's coding agent for the terminal, IDE and GitHub: understands your codebase, runs tasks and handles git
An environment for Claude Code that keeps context across sessions via hooks, ledgers and handoffs, orchestrates agents and saves tokens
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 Claude Code plugin marketplace with ready sets of agents, slash commands, hooks and skills, plus a search site
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| myclaude (this entry) | no | 2,747 | |
| BMad Method | no | 54,458 | the BMad process on its own, which here is just one of the modules |
| oh-my-claudecode | no | 40,169 | multi-agent orchestration inside Claude Code with an autopilot, without handing code to other backends |
| ZCF | no | 6,079 | a CLI that configures Claude Code and Codex in a single run |
README in short
The project description goes from a quick start to a table of modules with commands and links to each one's docs. Installation and update options follow, then module setup through config.json, picking a process by scenario, the orchestrator-and-executor role scheme and backend requirements. The end holds the ~/.claude layout after installation, troubleshooting tips, an FAQ and licensing terms.