blueprint

Turn a one-line objective into a step-by-step construction plan.

Updated Mar 19, 2026
One-click install
npx skills add https://github.com/RUSHYOP/imperium-cli --skill blueprint-rushyop
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blueprint
Source: https://github.com/RUSHYOP/imperium-cli/tree/main/content/skills/blueprint
Command: npx skills add https://github.com/RUSHYOP/imperium-cli --skill blueprint-rushyop

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Turn a one-line objective into a step-by-step construction plan for multi-session, multi-agent engineering projects. Each step has a self-contained context brief so a fresh agent can execute it cold. Includes adversarial review gate, dependency graph, parallel step detection, anti-pattern catalog, and plan mutation protocol.

Core Features & Use Cases

  • Cold-start execution: Every step includes a self-contained context brief, no prior context needed.
  • Adversarial review gate: Each plan is reviewed for completeness, dependencies, and anti-patterns.
  • Branch/PR/CI workflow: Generates branch and PR plans when git/gh are available; otherwise direct mode.
  • Parallel step detection: Builds a dependency graph to enable parallel execution where possible.
  • Plan mutation protocol: Supports splitting, inserting, skipping, reordering with audit trails.
  • Zero-runtime risk: Markdown-only skill compatible with Claude Code loader.

Quick Start

Provide a construction plan for a given objective by supplying a one-line goal.

Frequently Asked Questions about blueprint

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

FAQPage Schema
How do I turn a one-line objective into a step-by-step construction plan for multi-agent engineering?

To turn an objective into a construction plan, provide a one-line goal to generate a step-by-step blueprint with self-contained context briefs for each phase. This enables cold-start execution for fresh agents across multi-session projects without needing prior context.

What is a dependency graph for parallel step detection in multi-agent workflows?

A dependency graph for parallel step detection maps task relationships, enabling independent steps to execute simultaneously across multiple agents. This maximizes workflow efficiency by identifying which construction plan phases can run concurrently without conflicts.

Can I generate git branch and PR plans for multi-session engineering projects?

Yes, you can generate git branch and PR plans when git or gh workflows are available. The system operates in direct mode if these tools are absent, ensuring the construction plan remains compatible regardless of your version control setup.

Does the planning workflow support plan mutation like splitting or reordering steps?

Yes, the planning workflow supports plan mutation through splitting, inserting, skipping, and reordering steps. Every mutation is tracked with an audit trail, ensuring the dependency graph and step context remain consistent during multi-agent execution.

What is an adversarial review gate for engineering plan completeness?

An adversarial review gate validates engineering plan completeness by checking each drafted step for missing dependencies and anti-patterns. This verification occurs before registration, ensuring the final construction plan is robust enough for cold-start agent execution.

Do I need external dependencies to run the markdown-only blueprint workflow?

No external dependencies are required to run the markdown-only blueprint workflow. It operates with zero-runtime risk via the Claude Code loader, generating memory-indexed plans and step context briefs entirely through markdown formatting.