blueprint

Generate step-by-step construction plans for multi-agent engineering projects.

Updated Jul 4, 2026

Search Tags:#project management #parallel execution #multi-agent #step-by-step #construction plan

Ready to run

npx skills add https://github.com/Long-NguyenHai/warehouse-ai --skill blueprint-long-nguyenhai

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: blueprint
Source: https://github.com/Long-NguyenHai/warehouse-ai/tree/main/.agents/skills/blueprint
Command: npx skills add https://github.com/Long-NguyenHai/warehouse-ai --skill blueprint-long-nguyenhai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill simplifies the creation of detailed construction plans for multi-agent, multi-session engineering projects, ensuring a coherent and executable roadmap.

Core Features & Use Cases

Step-by-Step Plans: Generates comprehensive plans that break down complex objectives into actionable steps.
Adversarial Review: Ensures plans are complete, correct, and free of anti-patterns through a review process.
Parallel Execution: Detects parallel steps and assigns the appropriate model tier for efficient execution.
Use Case: Ideal for breaking down large features, refactors, or migrations that require coordination across multiple sessions or agents.

Quick Start

Generate a plan for your complex task by running /blueprint your-task-description.

Frequently Asked Questions about blueprint

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

FAQPage Schema

How do I create a step-by-step construction plan for a multi-agent engineering project?▼

To create a step-by-step construction plan for a multi-agent engineering project, use this Skill to break down complex objectives into actionable steps, handling multi-session coordination and dependency management automatically. Run the command to generate your roadmap.

What is adversarial review in multi-agent project planning?▼

Adversarial review in multi-agent project planning is a validation process that ensures generated construction plans are complete, correct, and free of anti-patterns before execution begins. It verifies the roadmap's coherence across multiple sessions.

Do I need Claude Code to generate multi-agent construction plans?▼

Yes, you need Claude Code to generate multi-agent construction plans, as it is a required dependency for full functionality. Git and GitHub CLI are optional dependencies that enhance parallel execution and multi-session coordination.

How does parallel execution work in multi-agent project plans?▼

Parallel execution in multi-agent project plans works by detecting independent steps within the construction plan and assigning the appropriate model tier for efficient execution. This optimizes resource usage across parallel agents.

Can I use this approach for large refactors and migrations across multiple sessions?▼

Yes, you can use this approach for large refactors and migrations requiring coordination across multiple sessions or agents. It simplifies complex engineering objectives into coherent, executable roadmaps with dependency management.

What are the limitations of automated construction plans for multi-agent projects?▼

Limitations of automated construction plans for multi-agent projects include the requirement for Claude Code to function properly. Without optional Git and GitHub CLI dependencies, full multi-session parallel execution and coordination features may be restricted.