From codex
Delegates sub-tasks to OpenAI Codex CLI for complex file editing, multi-step coding, or automated code generation. Supports both synchronous and background execution.
How this skill is triggered — by the user, by Claude, or both
Slash command
/codex:delegateThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use the `codex` MCP tools to delegate tasks to OpenAI Codex CLI.
Use the codex MCP tools to delegate tasks to OpenAI Codex CLI.
Codex must be installed and signed in:
npm i -g @openai/codex
# sign in via the Codex app — no API key needed
Use codex_run when you need the output before continuing:
codex_run(
prompt="<full task description with all context Codex needs>",
cwd="/path/to/project" # optional, defaults to current directory
)
Returns the final Codex message as a string.
Use codex_task_submit to fire off tasks and continue working:
Submit one or more tasks:
codex_task_submit(name="Refactor auth module", prompt="...", cwd="...")
codex_task_submit(name="Write tests for API", prompt="...", cwd="...")
Check status:
codex_task_status(task_id="<id>")
Fetch result when completed:
codex_task_result(task_id="<id>")
See all tasks:
codex_tasks_list()
| Variable | Purpose |
|---|---|
CODEX_PATH | Override path to the codex binary |
CODEX_SANDBOX | read-only, workspace-write, or danger-full-access — if unset, uses --dangerously-bypass-approvals-and-sandbox for headless operation |
npx claudepluginhub brainrot-creations/claude-plugins --plugin codexDelegates coding tasks (debug, implement, refactor) to OpenAI Codex CLI via codex exec, skipping the Node companion runtime for faster execution. Codex writes code; Claude verifies.
Delegates batch edits, boilerplate, multi-file refactors, and test scaffolding from Claude to Codex CLI to save tokens on mechanical work.
Executes Codex CLI for code analysis, refactoring, and automated edits with structured JSON output, file references, and sandbox controls.