blueprint

Convert a one-line objective into a five-phase construction plan.

Updated Apr 4, 2026

Search Tags:#automation #workflow #planning #multi-agent #blueprint #project-planning

Ready to run

npx skills add https://github.com/mitul-bhatia/Vibes --skill blueprint-mitul-bhatia

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: blueprint
Source: https://github.com/mitul-bhatia/Vibes/tree/main/.github/skills/blueprint
Command: npx skills add https://github.com/mitul-bhatia/Vibes --skill blueprint-mitul-bhatia

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Converts a single objective into a self-contained, multi-session construction plan that guides autonomous coding agents across multiple steps and sessions.

Core Features & Use Cases

Cold-start context: Each step includes a self-contained context brief so a fresh agent can execute it without prior memory.
Adversarial review gate: Every plan is reviewed by the strongest model against a checklist for completeness, dependency correctness, and anti-pattern detection.
Dependency graph and parallelism: The pipeline assigns serial and parallel execution order and tracks dependencies to optimize timelines.
Plan mutation protocol: Steps can be split, inserted, reordered, or deprecated with an audit trail.
End-to-end workflow: Registers the plan, updates memory index, and reports step counts and parallelism to the user.

Quick Start

Provide a one-line objective and Blueprint will generate a step-by-step construction plan.

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 multi-session construction plan for autonomous coding agents?▼

To turn a one-line objective into a multi-session construction plan, you provide your goal to generate a self-contained, step-by-step pipeline with explicit dependencies, parallelism rules, and cold-start context briefs for fresh agents.

What is multi-session project planning and when do I need it for large features?▼

Multi-session project planning tracks continuity of context across multiple PRs and teams. You need it for large features or migrations spanning multiple sessions where autonomous agents require reproducibility and guardrails against context loss.

Can I split, reorder, or insert steps in a multi-agent workflow plan after it is generated?▼

Yes, you can split, insert, reorder, or deprecate steps in a multi-agent workflow plan using the plan mutation protocol. This enforces an audit trail while allowing modifications to the serial and parallel execution order.

How does an adversarial review gate check dependency correctness in multi-step project plans?▼

An adversarial review gate checks dependency correctness by reviewing every generated plan against a completeness checklist. The strongest model detects anti-patterns and validates serial and parallel execution rules before the plan is registered.

Does multi-session planning work without prior agent memory or context?▼

Yes, multi-session planning works without prior agent memory because each step includes a self-contained context brief. This cold-start context ensures a fresh autonomous coding agent can execute the step without previous session memory.

What is the best way to prevent context loss when automating workflows across multiple PRs?▼

The best way to prevent context loss when automating workflows across multiple PRs is enforcing a five-phase pipeline with explicit dependencies and cold-start briefs. This ensures continuity, tracks parallelism, and updates the memory index end-to-end.