deep-plan

Convert vague feature requests into validated implementation plans via research, interviews, and multi-LLM review.

104|19|Updated Jan 15, 2026
One-click install
npx skills add https://github.com/piercelamb/deep-plan --skill deep-plan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-plan
Source: https://github.com/piercelamb/deep-plan/tree/main/skills/deep-plan
Command: npx skills add https://github.com/piercelamb/deep-plan --skill deep-plan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Converts vague feature requests into concrete, validated implementation plans by orchestrating research, stakeholder interviews, and multi-LLM review.

Core Features & Use Cases

  • End-to-end planning orchestration: Research → Interview → External LLM Review → TDD Plan to produce production-ready guidance.
  • Spec-driven workflow: Synthesize initial spec, research findings, and interview insights into a cohesive plan.
  • Guardrails and traceability: Environment validation, session management, and artifact generation with clear audit trails.

Quick Start

Run /deep-plan @path/to/your-spec.md to begin an AI-assisted planning workflow.

Frequently Asked Questions about deep-plan

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

FAQPage Schema
How do I convert a vague feature request into a concrete implementation plan?

End-to-end AI-assisted planning converts vague feature requests into concrete implementation plans by orchestrating research, stakeholder interviews, and multi-LLM external review into production-ready guidance.

How do I start an AI-assisted planning workflow for complex features?

Run /deep-plan @path/to/your-spec.md to start the AI-assisted planning workflow, which synthesizes your initial spec, research findings, and interview insights into a cohesive TDD plan.

Do I need Python and external LLM access to run multi-LLM feature review?

Yes, multi-LLM external review requires Python 3.11+ with uv installed, plus access to Gemini or OpenAI APIs to validate your plans and generate planning artifacts like claude-plan.md.

What is the best way to structure stakeholder interviews and research for TDD planning?

Spec-driven workflows structure stakeholder interviews and research by synthesizing findings into cohesive plans, using environment validation and session management to maintain clear audit trails and guardrails.

When should I not use automated end-to-end feature planning?

Automated end-to-end feature planning should be avoided for simple tasks that lack complex research requirements or do not benefit from structured stakeholder interviews and multi-LLM external review.