deep-planning

Enforce plan-first workflows with phase-based investigation and evidence-backed verification.

Updated Mar 6, 2026
One-click install
npx skills add https://github.com/guitardaveews/deep-planning-skill --skill deep-planning-guitardaveews
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-planning
Source: https://github.com/guitardaveews/deep-planning-skill/tree/main
Command: npx skills add https://github.com/guitardaveews/deep-planning-skill --skill deep-planning-guitardaveews

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep Planning mandates deep planning and mandatory investigation before any implementation task is executed.

Core Features & Use Cases

  • Enforces plan-first workflows and explicit problem understanding
  • Leverages subagent strategy to offload research and keep context clean
  • Provides verification-before-done with audits, logs, and evidence
  • Includes risk/scope controls like scope-boundary-detection and dependency mapping

Quick Start

Enter plan mode immediately and start Phase 0 by listing unknowns, mapping data flow, and drafting a detailed plan.

Frequently Asked Questions about deep-planning

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

FAQPage Schema
How do I enforce investigation before implementation in complex software tasks?

Deep planning prevents architectural errors by requiring mandatory investigation, phase-based research, and evidence-backed verification before executing multi-file changes or complex software tasks.

What is the best way to manage risk and scope boundaries in multi-file changes?

Risk and scope boundaries in multi-file changes are managed through structured task management artifacts that map dependencies, detect scope boundaries, and require verification-before-done with logs and audits.

How do I start a plan-first workflow for architectural decisions?

A plan-first workflow begins by entering plan mode and executing Phase 0 to list unknowns, map data flow, and draft a structured plan, keeping context clean through subagent research offloading.

Does deep planning work for team coordination on complex software tasks?

Deep planning facilitates team coordination by enforcing phase-based investigation and producing structured plans and lessons, ensuring complex software tasks are verified before being marked done.

When should I not use a mandatory investigation workflow?

Avoid mandatory investigation workflows for simple tasks where the overhead of phase management, scope-boundary-detection, and evidence-backed verification outweighs the complexity of the implementation itself.