TeamFlow in OpenAI Codex CLI

1. Install

npx -y github:macleodlabs-ai/teamflow-plugin skills install --for codex

That writes:

• .agents/skills/<skill>/SKILL.md — the twelve skills, in the shared skills directory this tool reads
• AGENTS.md — the always-on reporting instruction, as a marked block; your own instructions above and below it are kept
• ~/.codex/config.toml — the TeamFlow MCP server, as a marked TOML table
• .teamflow/hooks/codex.sh — the hook itself; it calls teamflow hook --for codex, prints nothing the tool could read as a denial, and exits 0 whatever happens
• .codex/hooks.json — the hook registration: SessionStart, UserPromptSubmit, PermissionRequest, PostToolUse, SubagentStart, SubagentStop, PreCompact, Stop and SessionEnd

Add --scope user to install for every project on this machine, or --dry-run to see the file list without writing anything. Re-running it replaces TeamFlow's block and leaves the rest of each file alone.

Then sign this machine in, and check it. The hooks above report nothing at all until there is a credential, and nothing errors while they do not: an install that is never signed in looks exactly like a working one until somebody notices the board is empty.

npx -y github:macleodlabs-ai/teamflow-plugin login
npx -y github:macleodlabs-ai/teamflow-plugin status

login opens a browser once and stores a revocable refresh token; login --device prints a code to type into a browser on another machine instead. status answers with the organisation, the project and the repository this checkout reports to, and identity: not signed in there is the one thing to fix before expecting anything on the board.

2. What TeamFlow can do in Codex

This list comes from Codex's own documentation, checked 2026-10-04. We have not run Codex with TeamFlow yet.

• Yes: Ticket, project and lead notes at session start.
• Yes: Ticket, project and threads with each prompt.
• Yes: Allow or Deny from TeamFlow. Codex has no event for a question with options, so you answer those in Codex.
• Yes: Plan goes on when a turn ends.
• Yes: Helper agents and session end on the board.
• Yes: TeamFlow skills.
• Yes: TeamFlow MCP server.
• No: One-minute check-in on an idle session. Codex cannot wake an idle session later.

To add these to an earlier install, run npx -y github:macleodlabs-ai/teamflow-plugin hooks install --for codex again. Every tool side by side.

3. The skills you get

The same twelve the Claude Code plugin carries, generated from the same SKILL.md files. This tool loads them as skills, one directory each, with the name prefixed so they do not collide with anybody else’s in a shared directory.

Each one wraps a github:macleodlabs-ai/teamflow-plugin subcommand, so the skill and the shell command are the same thing. Inside Claude Code they are /teamflow:login, /teamflow:status and so on.

4. Sign in once

codex mcp login teamflow

That opens the browser for an OAuth 2.1 authorization-code flow with PKCE, on a loopback callback. codex mcp list then shows teamflow with its three tools.

5. By hand

The installer writes this for you. If you would rather paste it: ~/.codex/config.toml, or .codex/config.toml in a trusted project.

[mcp_servers.teamflow]
url = "https://codercat.io/mcp"

Or let the CLI write it:

codex mcp add teamflow --url https://codercat.io/mcp

auth defaults to oauth, so nothing else is needed. Ignore guides that tell you to set experimental_use_rmcp_client; that flag was for builds that ignored remote servers without it and is not part of the current configuration.

There is no .codex/skills. .agents/skills is Codex's only skill location, searched in the working directory, its parents up to the repository root, and $HOME. Anything written elsewhere is invisible to it.

--scope user puts the skills in ~/.agents/skills instead, which is the right choice for a machine where every repository should have them. That directory is not Codex's alone: Cursor, Copilot, Gemini CLI, Zed, Windsurf and Junie all read it, so one user-scope install covers every one of them.

6. Uninstall

npx -y github:macleodlabs-ai/teamflow-plugin skills uninstall --for codex

That removes exactly what the install above wrote: its own files go, and its own entries come out of the files it shares with your editor's other hooks and servers, which are left as they were. A hook TeamFlow did not write is never touched, and uninstalling something that is not there succeeds quietly. --dry-run prints the list without removing anything.

npx -y github:macleodlabs-ai/teamflow-plugin hooks uninstall --all is the bigger hammer: every tool at once, the git hooks, and this repository's ticket binding.

7. The stages to report

BACKLOG is the thirteenth: the ticket exists and nothing has started. Statuses are running, success, waiting, blocked, failed and idle. This is the same vocabulary the Claude Code plugin derives, so a ticket moved from here sits in the same column as one moved by Claude.

8. Without the agent

Anything that runs in a shell posts the same report with no model involved. A git hook, an npm script, a run configuration, a CI step:

npx -y github:macleodlabs-ai/teamflow-plugin report --issue DAEMON-142 --stage LOCAL_TEST \
  --status success --summary "42 tests, 0 failing"

Same envelope, same transport, same credential. Prefer it wherever the board has to be right whether or not an agent remembered. npx -y github:macleodlabs-ai/teamflow-plugin report --help lists every flag.

9. What this is not

OpenAI Codex CLI reports automatically through its own hooks: SessionStart, UserPromptSubmit, PermissionRequest, PostToolUse, SubagentStart, SubagentStop, PreCompact, Stop and SessionEnd. Nothing to remember. Codex adopted Claude Code’s hook shape field for field, so TeamFlow reports from it as it does from Claude Code: context at the start and on each prompt, answers to permission requests from TeamFlow, and plan work that carries on at a stop. Checked against the vendor’s own hook documentation on 2026-10-04: Codex lists twelve events: SessionStart, SessionEnd, PreToolUse, PermissionRequest, PostToolUse, PreCompact, PostCompact, UserPromptSubmit, SubagentStart, SubagentStop, Stop and Interrupt, and "Hooks are enabled by default." SessionStart and UserPromptSubmit take "hookSpecificOutput.additionalContext – Added as developer context". PermissionRequest takes hookSpecificOutput.decision.behavior "allow", or "deny" with message, and "any deny wins". On Stop, decision: "block" "tells Codex to continue". There is no asyncRewake and no hook for a question the agent asks the person, so the one-minute check-in and answered questions stay with Claude Code. "Non-managed hooks require explicit review before execution. Use the /hooks command".

The hooks see this tool's own calls and nothing else. A test run in another terminal, a merge from the web UI, a deploy from a pipeline: put npx -y github:macleodlabs-ai/teamflow-plugin report in the git hook or the script that already runs on every change, and those land on the board too.

10. What leaves your machine

TeamFlow manages your AI software factory. It is run by MacleodLabs. Your code stays inside your organisation: the local plugin sends derived state only, a ticket key, a stage, a status, a short summary and evidence links. Prompts, diffs, source code, commands and logs are never sent and are rejected before they are charged.

• Every client
• llms.txt
• Capabilities
• The reporting contract, in full