mylukin

Break down PRD into atomic, testable tasks using CLI for modular storage

Autonomous implementation loop with TDD, fresh agents per task, and self-healing

Interactive requirement clarification through structured questions and PRD generation

Autonomous language and framework detection for any programming language. Use when user asks to "detect language", "what language is this project", or when initializing ralph-dev for a new project.

Autonomous end-to-end development from requirement to delivery. Use when user wants complete automation, "build X for me", or full feature implementation without manual steps.

Systematic error recovery using root cause investigation before fixes

Two-stage code review, quality gates, and automated delivery (commit + PR)

Implements a single task following the next → implement → check → done workflow with TDD support. Use when working on one specific task, implementing a single feature from the backlog, or following TDD red-green-refactor cycle. Triggers on 'next task', 'next feature', 'implement feature', 'work on feature', 'single task mode', 'what should I work on'.

Executes unattended batch processing of all pending tasks with autonomous decision-making. Use when running all tasks automatically, batch processing without supervision, completing entire feature backlog, or working on a specific task by ID. Triggers on 'run all tasks', 'complete all features', 'batch processing', 'unattended mode', 'auto-complete tasks', 'run feature'.

Multi-role requirement analysis and task breakdown workflow using 4 specialized AI agents (PM, UX, Tech, QA). Each agent conducts web research before analysis to gather industry best practices, case studies, and current trends. Supports Quick Mode (parallel, ~3 min, one Q&A session) and Deep Mode (serial, ~8 min, Q&A after EACH agent so answers inform subsequent analysis). Triggers on 'foreman-spec', 'spec feature', 'break down requirement', 'define tasks', 'spec this'.

Creates AI agent task management structure with feature backlog (ai/tasks/), TDD enforcement, and progress tracking. Use when setting up agent-foreman, initializing feature-driven development, creating task backlog, or enabling TDD mode. Triggers on 'init harness', 'setup feature tracking', 'create feature backlog', 'enable strict TDD', 'initialize agent-foreman'.

Scans codebases to generate architecture documentation (ARCHITECTURE.md). Use when joining an existing project, understanding codebase structure, exploring project architecture, or preparing for agent-foreman init. Triggers on 'analyze project', 'understand codebase', 'explore architecture', 'scan project structure', 'survey project'.

Delegate coding tasks to Codex, Claude Code, or Pi agents via background process. Use when: (1) building/creating new features or apps, (2) reviewing PRs (spawn in temp dir), (3) refactoring large codebases, (4) iterative coding that needs file exploration. NOT for: simple one-liner fixes (just edit), reading code (use read tool), thread-bound ACP harness requests in chat (for example spawn/run Codex or Claude Code in a Discord thread; use sessions_spawn with runtime:"acp"), or any work in ~/clawd workspace (never spawn agents here). Claude Code: use --print --permission-mode bypassPermissions (no PTY). Codex/Pi/OpenCode: pty:true required.

Maintain the OpenClaw memory wiki vault with deterministic pages, managed blocks, and source-backed updates.

End-to-end Parallels smoke, upgrade, and rerun workflow for OpenClaw across macOS, Windows, and Linux guests. Use when Codex needs to run, rerun, debug, or interpret VM-based install, onboarding, gateway smoke tests, latest-release-to-main upgrade checks, fresh snapshot retests, or optional Discord roundtrip verification under Parallels.

Run, watch, debug, and extend OpenClaw QA testing with qa-lab and qa-channel. Use when Codex needs to execute the repo-backed QA suite, inspect live QA artifacts, debug failing scenarios, add new QA scenarios, or explain the OpenClaw QA workflow. Prefer the live OpenAI lane with regular openai/gpt-5.4 in fast mode; do not use gpt-5.4-pro or gpt-5.4-mini unless the user explicitly overrides that policy.

Get multi-model AI consensus on complex questions using Claude, Codex, and Gemini