blueprint

Generate multi-step construction plans with phase-based design and dependency edges.

1|Updated Mar 19, 2026
One-click install
npx skills add https://github.com/devs6186/claude-private-skills-agents-commands --skill blueprint-devs6186
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blueprint
Source: https://github.com/devs6186/claude-private-skills-agents-commands/tree/main/skills/blueprint
Command: npx skills add https://github.com/devs6186/claude-private-skills-agents-commands --skill blueprint-devs6186

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Turn a one-line objective into a step-by-step construction plan that any coding agent can execute cold.

Core Features & Use Cases

  • Cold-start execution — Every step includes a self-contained context brief.
  • Adversarial review gate — Every plan is reviewed for completeness, dependencies, and safety.
  • Branch/PR/CI workflow integration — Plans include workflow considerations; direct mode if CI is unavailable.
  • Parallel step detection — Identifies steps that can run concurrently to speed delivery.
  • Plan mutation protocol — Steps can be split, inserted, reordered, or abandoned with audit trails.
  • End-to-end plan generation — Produces a complete, auditable plan ready for execution across agents.

Quick Start

Provide a concise objective and run the /blueprint command to generate a detailed, multi-step construction plan for your project.

Frequently Asked Questions about blueprint

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

FAQPage Schema
How do I generate a multi-agent project plan from a single objective?

Multi-agent project plan generation starts with a concise objective and runs through a 5-phase workflow (research, design, draft, review, register) to produce detailed, step-by-step construction plans with dependency edges and rollback strategies.

What is the best way to coordinate parallel tasks across multiple coding agents?

Coordinating parallel tasks across coding agents requires phase-based design that identifies parallelizable steps and includes self-contained context briefs, ensuring sub-agents execute concurrently without dependency conflicts.

Can I modify or reorder steps in an existing multi-agent software engineering plan?

Steps in a multi-agent software engineering plan can be split, inserted, reordered, or abandoned using a plan mutation protocol that maintains complete audit trails for traceability and accountability.

How does the adversarial review process work for multi-agent construction plans?

The adversarial review process evaluates every generated multi-agent construction plan for completeness, dependencies, and safety by acting as a gate before final registration, ensuring auditable outputs ready for cold-start execution.

Do I need a CI pipeline to use multi-agent plans for software project management?

A CI pipeline is not strictly required for multi-agent software project management plans; the system includes workflow considerations for Branch/PR/CI integration but operates in direct mode if CI is unavailable.

Why does multi-agent plan generation require a memory index and Markdown output?

Multi-agent plan generation requires Markdown output written to a plans directory and a memory index to track progress across phases, ensuring end-to-end traceability and persistent state for sub-agent coordination.