blueprint

Generate self-contained step plans with dependency graphs for multi-session engineering projects.

Updated Mar 26, 2026
One-click install
npx skills add https://github.com/luongldptit/move-ticket --skill blueprint-luongldptit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blueprint
Source: https://github.com/luongldptit/move-ticket/tree/main/.agent/skills/blueprint
Command: npx skills add https://github.com/luongldptit/move-ticket --skill blueprint-luongldptit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Large, multi-session engineering projects such as multi-PR feature builds, cross-session refactors, or cross-agent workstreams often suffer from context loss between sessions, leading to rework and inconsistent execution when fresh agents or team members pick up work without prior context.

Core Features & Use Cases

  • Step-by-step construction plans: Breaks one-line project objectives into 3-12 one-PR-sized steps with clear dependency ordering, parallel step detection, and per-step rollback strategies.
  • Cold-start execution: Every step includes a self-contained context brief, task list, verification commands, and exit criteria so any agent can execute it without reading prior steps.
  • Adversarial review gate: All plans are reviewed by a strongest-model sub-agent against a checklist and anti-pattern catalog to catch completeness, dependency, and design issues before finalization. Use case: Plan a multi-session PostgreSQL database migration or a cross-agent LLM provider plugin system implementation with clear parallel workstreams.

Quick Start

Use the blueprint skill to generate a complete, executable construction plan for your multi-session engineering project objective.

Frequently Asked Questions about blueprint

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

FAQPage Schema
How do I plan multi-PR feature development across multiple engineering sessions?

Multi-PR feature development requires breaking a project objective into 3-12 one-PR-sized steps with dependency ordering and parallel workstream detection. This generates self-contained step plans with context briefs and exit criteria so any agent can execute without prior project context.

What is cold-start execution context in cross-session code refactoring?

Cold-start execution context provides a self-contained brief, task list, verification commands, and exit criteria for each step. This allows any fresh agent or team member to execute cross-session code refactors consistently without reading prior steps or needing previous project context.

How to coordinate parallel workstreams across sub-agents for software engineering projects?

Parallel workstreams across sub-agents require step-by-step construction plans with clear dependency graphs and parallel step detection. Each step includes per-step rollback strategies and adversarial review gates to catch completeness and design issues before finalization.

When do I need an adversarial review gate for project planning?

An adversarial review gate is needed during project planning when finalizing multi-step engineering plans. A strongest-model sub-agent reviews the plan against a checklist and anti-pattern catalog to catch dependency, completeness, and design issues before execution begins.

Can I use dependency graphs for cross-agent workstream coordination?

Dependency graphs enable cross-agent workstream coordination by mapping 3-12 one-PR-sized steps with clear dependency ordering. This allows parallel workstream detection and per-step rollback strategies, ensuring consistent execution across multiple agents without context loss.

What are the limitations of manual planning for multi-session engineering projects?

Manual planning for multi-session engineering projects often suffers from context loss between sessions, leading to rework and inconsistent execution. Without self-contained step plans and cold-start execution context briefs, fresh agents picking up work lack the necessary prior project context for consistent delivery.