codex-cto-agent-team skill
A skill that hands work to a team of a CTO, five PMs and fifty engineers: roles, permissions, a task log and resource leases.
- 86 ★ on the repository
- Last commit
Extended access
Why this level
- The skill organizes the creation of many agent chats, each spending account limits
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/codex-cto-agent-teamDetects 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 skill organizes the creation of many agent chats, each spending account limits
This is third-party code. Review the repository files before installing.
What it does
The `cto-agent-team` skill defines an organization of 56 roles: a CTO, five product managers (interface, services and data, systems and devices, automation, quality and release) and ten engineer roles under each. The CTO takes in work, hands it to PMs and accepts the result, PMs divide tasks and review results and engineers return outputs and evidence. A task gets a target, permissions, outputs, acceptance criteria and writable resources. State lives in a private folder, resources are pinned to one writer and extra messages are avoided: PMs read shared events. Model routes are described in the `compatibility-6.1` profile. Documentation exists for Claude Code and Cursor but they have no live client validation.
Who it is for. Leads of large Codex assignments who need a steady role structure instead of one agent and want plans with task owners and criteria.
Good fit when
- A large project must be broken down by task owners and acceptance criteria without creating chats.
- Five PM and fifty engineer chats are created with explicit user permission.
- A shared resource is edited by one executor while others read, to avoid conflicts.
- Model routes by kind of work are to be described explicitly.
Not a fit when
- The task is small: an organization of 56 roles loads it with overhead.
- The host cannot create chats and messages between them: the skill does not add those abilities.
- Live work in Claude Code or Cursor is needed: only guidance exists for them, with no validation.
Example request
Use $cto-agent-team to plan the work for the goal, give the task owners and acceptance criteria. Do not make chats.Limitations
The skill adds no host abilities: creating chats, messages between them and monitoring depend on supported tools and account limits. A role definition does not create a chat and a created chat, a READY reply and active computation stay different states. Claude Code and Cursor have documentation with no validation on live clients. The model profile `compatibility-6.1` is called explicit and claims no model availability. Python 3.11 with only the standard library is needed. The repository has no license file.
How to disable. Delete the `cto-agent-team` folder from the skills directory. Delete private task and state records in your chosen folder separately.
FAQ
How do I install it?
Copy the whole `skills/cto-agent-team` folder into the personal skills directory `~/.agents/skills/` or the project `.agents/skills/` and restart Codex.
Does selecting the skill permit creating chats?
No. Chat creation and messages must be permitted explicitly in the request, otherwise the skill only builds a plan.
How many roles are in the organization?
Fifty-six: the CTO, five PMs and fifty engineers, including the current CTO chat.
How does the skill work with Antigravity?
Register the Codex or `~/.agents/skills` directory in `~/.gemini/config/skills.json` under `entries`.
What do the local helpers need?
Python 3.11 or newer and only the standard library.
Discussion
No questions or reviews yet.
Related
SkillsA Claude Code skill that hands implementation-heavy work to Codex CLI while Claude keeps planning, review and acceptance.
A skill that links two to four open Codex CLI windows: one splits the work while the others execute, cross-check and deliver.
A Codex skill that splits work across three models: the main thread plans, Terra handles the screen, Luna runs the tests.
Team Mode splits a Codex task between an explorer, an executor and a reviewer while the main agent keeps final acceptance
A desktop app where Claude Code, Codex, Gemini and other agents work as a team of specialists in a shared project
One agent launches others in the background: Claude Code, Codex, Copilot CLI, Grok Build and Pi, with a model chosen per stage.
README in short
The README shows the organization chart, installation for Codex, example prompts, model routes and private state. Details live in the skill references.