buff-plan

Generates architecture-driven project plans and writes them to .buff/plan.md.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/qkal/claude-beta-script --skill buff-plan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: buff-plan
Source: https://github.com/qkal/claude-beta-script/tree/main/skills/plan
Command: npx skills add https://github.com/qkal/claude-beta-script --skill buff-plan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Pattern-informed adaptive planning that reads from counterfeit memory, eliminates redundant questions, and writes versioned plans to .buff/plan.md. Supports --refine, --diff, and --version commands.

Core Features & Use Cases

  • Generates architecture-driven project plans by scanning the repository and existing memory, writing to .buff/plan.md.
  • Supports refine/diff/version workflows to evolve plans over time.
  • Writes decisions to counterfeit memory and records patterns for memory-driven planning.

Quick Start

Run /buff:plan with a goal to generate a versioned project plan.

Frequently Asked Questions about buff-plan

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

FAQPage Schema
How do I generate a versioned architectural plan for a new software project?

Yes, the planning tool supports existing codebase extensions. It reads from memory to eliminate redundant questions and generates architecture-driven plans that adapt to your current repository structure.

How does memory-driven planning work for software architecture?

Use the refine command to adapt existing plans. The diff command compares plan versions, and the version command manages versioning, allowing your architectural plan to evolve over time.

Can I refine or diff a project plan after it has been generated?

Yes, the planning tool supports refine, diff, and version workflows. You can use these commands to evolve plans over time, compare changes, and manage version history within the .buff/plan.md file.

Do I need to manually configure memory stores before generating an architecture plan?

No manual configuration is required. The planning tool automatically coordinates context from memory, reads existing patterns to eliminate redundant questions, and writes architectural decisions back to memory stores.

What is the best way to extend an existing codebase with a new architectural plan?

The best way to extend an existing codebase is to trigger a planning session. It scans your repository, reads stored memory patterns, and writes a versioned architectural plan to .buff/plan.md.