great_cto plugin
A Claude Code plugin that runs a task through stages with review and gates, with a local board of decisions and execution.
- 115 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- Agents write code and run commands in the project, and the devops role can deploy once gates pass.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/great-ctoDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Agents write code and run commands in the project, and the devops role can deploy once gates pass.
This is third-party code. Review the repository files before installing.
What it does
great_cto coordinates Claude Code work and experimental controlled Codex runs through explicit stages, review and gates. The /start command takes a new product or a task in an existing project and picks the workflow itself, /inbox shows what awaits your decision and /resume continues only what is approved. In the terminal the same steps are great-cto run, status and resume. The great-cto board command opens a local board at localhost:3141: tasks with acceptance criteria, pending decisions, project documents and diagrams, and token consumption. Missing evidence stays unknown and does not count as a passed check. There are eight core roles, from product owner and architect to reviewer, QA, security officer and devops, plus required project reviewers, and the full catalog exceeds seventy agents.
Who it is for. Engineering leads and solo developers who want tasks, checks and agent decision points on one board.
Good fit when
- A new feature is taken on and a pipeline with independent review is needed before building.
- You need to see which agent decisions wait for a human and which are already made.
- Token consumption per project must be visible separately from estimates and budgets.
- A web release goes through build, testing and approval before deployment.
Not a fit when
- You need a simple helper for one edit: the role pipeline targets larger tasks.
- You need a ready fully autonomous Codex: the controlled mode is experimental.
- Agents with code and deployment rights must not run without manual oversight.
Example request
/start "add Google login" and show in /inbox which decisions wait for me before the build.Limitations
The README warns of a verified-distribution boundary: in the current implementation init, plugin refresh and companion installation refuse activation, and the listed commands describe the earlier 3.70.1 flow. The repository is about 200 MB. For Claude Code the whole chain works, while for Codex the controlled mode is experimental and needs both CLIs installed and authenticated. A catalog of agents is not proof of execution or expertise. Costs go to your provider account.
How to disable. Remove the plugin from Claude Code with /plugin uninstall and delete the .great_cto directory from the project if you no longer need it. Close the local board by stopping the great-cto board process.
FAQ
How do I start a task?
In Claude Code enter `/start "add Google login"`. In the terminal `great-cto run "add Google login"` does the same.
Where do I see the board?
Run `great-cto board`: it opens locally, usually at `localhost:3141`.
What stops the pipeline?
The `approval-level` in `.great_cto/PROJECT.md` sets which decisions make agents wait for you.
Does great_cto work with Codex?
Yes, in three modes: the native Claude Code pipeline, an experimental controlled Codex run and the Codex plugin in an ordinary session with skills, MCP and safety hooks.
Does the project charge anything?
No. The MIT-licensed project bills nothing, and costs go to your provider account.
Discussion
No questions or reviews yet.
Related
PluginsA plugin that readies a project for Claude Code: guidance, tests, review, TDD, refactoring and a resumable fix loop
Claude Code plugin with 17 skills: task sizing, design review, TDD, fresh-context subagents and a PR after three human decisions.
A Claude Code plugin for development with separated roles: spec, review, TDD, QA and verification by different agents
A universal workflow plugin for Claude Code: 7 agents, 26 commands, 14 skills and 14 hooks for any project
A Claude Code plugin: Claude does the task, then a stop hook starts an independent Codex review and Claude fixes the findings in the same session
A Claude Code plugin: 39 commands for PRs, review, commits and specs, nine expert roles and hooks, in eight language versions
README in short
The project presents itself as a board around an agent you already have: tasks, recorded execution, documents and consumption in one workspace. The README separates execution, evidence and unknown values explicitly. Documentation is translated into ten languages, including Russian.