ceo-ralph

Cancel active execution loop and cleanup state

Start, stop, or open the CEO Ralph web dashboard

Generate technical design from requirements

Generate a single discovery.md that merges research and requirements.

One-shot Remote UI + HTTPS tunnel setup

This is an alias for `/ceo-ralph:implement`.

Submit feedback or report an issue for CEO Ralph plugin.

Show help and available commands

Start task execution loop using the best available executor (Codex MCP or Task sub-agents).

Create new spec and start discovery phase

Pause the execution loop

Generate a merged execution plan: a short design summary plus tasks.md.

You are helping a user update their specification files after execution. This command provides a methodical, file-by-file, section-by-section approach to updating specs.

Configure and start the Remote UI for approvals

Generate requirements from goal and research

Run or re-run research phase for current spec

Resume a paused execution

Configure executor (Codex MCP or Task tool fallback) and verify setup

Smart entry point for CEO Ralph. Detects whether to create a new spec or resume existing one.

Show rich status with worker tree, progress bars, and delegation stats

Switch active spec

Generate implementation tasks from design

Remove CEO Ralph MCP configuration

Expert systems architect for technical design. Masters system design, component architecture, patterns, and technical trade-off analysis.

Main orchestration brain that coordinates all phases and reviews all outputs

Reviews Codex worker outputs against acceptance criteria and quality standards

Synthesizes merged spec artifacts in one pass for quick mode. Generates discovery.md and tasks.md with minimal user interaction.

Expert product manager for requirements gathering. Focuses on user stories, acceptance criteria, business value, and user-centric development.

QA engineer that runs verification commands and checks acceptance criteria for [VERIFY] tasks.

Expert at methodically updating spec files after execution. Reviews sections, gathers feedback, and updates specifications incrementally.

Expert analyzer and researcher that never assumes. Always verifies through web search, documentation, and codebase exploration before providing findings. Use for initial project research, feasibility analysis, and gathering context before requirements.

Autonomous task executor for spec-driven development. Executes a single task from tasks.md, verifies, commits, and signals completion. Invoked via Codex MCP or Claude Task sub-agent.

Expert task planner for breaking design into executable tasks. Masters POC-first workflow, task sequencing, and quality gates.

How to delegate implementation tasks to workers via Codex MCP or Task sub-agents. Use when executing tasks that require code implementation.

Output rules for all agents - concise, scannable, actionable. Based on Matt Pocock's planning principles.

Core principle that the main agent is a coordinator, not an implementer. All work must be delegated to subagents.

Detect fix vs add goals, diagnose failures before work, verify fixes after. Ensures "fix X" specs actually fix X.

Criteria for reviewing Codex worker outputs across 5 dimensions. Use when reviewing code submitted by workers before approval.

Spec-driven development workflow for building features with research, requirements, design, and task phases

4-layer verification system for task completion. Use when verifying that a task has been fully completed before marking it done.

3 hooks across 3 events