skills.rest
bosmadev
bosmadevCOMMUNITY
Explore all skills by this author →
GitHub

start

Plan and execute tasks by spawning parallel agents under Ralph mode.

Updated Aug 23, 2026
Search Tags:#workflow#planning#multi-agent#safety#autonomous#start#ralph
One-click install
npx skills add https://github.com/bosmadev/claude --skill start-bosmadev
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: start
Source: https://github.com/bosmadev/claude/tree/main/skills/start
Command: npx skills add https://github.com/bosmadev/claude --skill start-bosmadev

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates complex, multi-step development tasks by orchestrating parallel agents under Ralph Mode to plan, implement, and review work with planned guardrails.

Core Features & Use Cases

  • ULTRATHINK planning: Deep analysis and automatic task decomposition across multiple agents.
  • Parallel execution: Simultaneous implementation and review to accelerate delivery.
  • Model routing & safety: Built-in governance, session tracking, and gating to ensure verifiable progress.

Quick Start

Invoke /start to launch Ralph Mode with default settings and auto-loop.

Frequently Asked Questions about start

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I plan and execute multi-step development tasks autonomously?▼

Autonomous multi-step task execution is handled by spawning parallel agents under Ralph mode. It uses ULTRATHINK planning and multi-model routing across Sonnet and Opus to decompose, implement, and review work with safety gates.

What is ULTRATHINK planning for multi-agent workflows?▼

ULTRATHINK planning is a deep analysis mechanism that automatically decomposes complex development tasks across multiple parallel agents. It ensures structured task distribution and creates plan artifacts before implementation begins.

How do I start a Ralph mode workflow with parallel agents?▼

You can start a Ralph mode workflow by invoking the /start command. This launches the multi-agent environment with default settings, auto-loop capabilities, and session tracking for safe task completion.

Does multi-agent autonomous execution support model routing between Sonnet and Opus?▼

Yes, multi-agent autonomous execution supports multi-model routing between Sonnet and Opus. This allows dynamic model assignment during parallel implementation and optional post-review workflows across import, noreview, and review scenarios.

What are the limitations of autonomous parallel agent execution for software engineering?▼

Limitations include dependency on Ralph state persistence and mandatory safety gates. While parallel execution accelerates delivery, all progress must pass enforced governance, task lists, and session tracking to ensure verifiable completion.

Related Skills

Explore More Software Engineering Skills→
TienchingTienching
3

ralph-init

Generate a canonical PRD and initialize Ralph progress tracking.

Community
Intermediate
FFCfelps1FFCfelps1

ce-work

Plan, execute, and ship software features with code review and CI/CD integration.

Community
Advanced
Xza85hrfXza85hrf

cost-aware-llm-pipeline

Route tasks across Opus, Sonnet, Haiku, Ollama Cloud, and Ollama Local to minimize token costs.

Community
Intermediate
hainamchunghainamchung
54

agent-assistant-auto

Orchestrate Codex-native agents to execute autonomous task workflows.

Community
Advanced