rpi-plan

Creates phased implementation plans through interactive codebase research and iterative user feedback.

5|Updated Sep 24, 2017
One-click install
npx skills add https://github.com/madyankin/dotfiles --skill rpi-plan-madyankin
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rpi-plan
Source: https://github.com/madyankin/dotfiles/tree/main/.config/agents/skills/rpi-plan
Command: npx skills add https://github.com/madyankin/dotfiles --skill rpi-plan-madyankin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Writing implementation plans without understanding the existing codebase leads to missed constraints, scope creep, and rework. This Skill structures the planning process so every plan is grounded in actual code, verified research, and explicit user alignment before implementation begins. ## Core Features & Use Cases - Codebase-Grounded Research: Spawns parallel research agents to locate relevant files, analyze current implementations, and find existing patterns before proposing any design. - Iterative Plan Development: Moves through context gathering, research, outline feedback, and detailed writing phases, pausing for user confirmation at each stage. - Phased Plan Documents: Writes timestamped plans to _scratchpad/plans/ with per-phase changes, code snippets, and separated automated versus manual verification criteria. - Use Case: You receive a ticket to add authentication to an API. The Skill reads the ticket, researches the existing middleware and routing code, presents design options with trade-offs, then writes a phased plan with testable success criteria for each phase. ## Quick Start Ask the agent to create an implementation plan for your feature or ticket, for example: plan the implementation of the password reset feature described in this ticket.

Frequently Asked Questions about rpi-plan

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

FAQPage Schema
How do I create an implementation plan before coding a feature?▼

Provide the ticket or feature description and let the Skill read all referenced files, then spawn parallel research agents to analyze the codebase. It presents findings and design options for your approval before writing a phased plan with verification criteria.

What should a good technical implementation plan include?▼

A solid plan includes current state analysis, desired end state, explicit out-of-scope items, phased changes with file-level detail and code snippets, plus separated automated and manual success criteria. This Skill generates plans following exactly that structure.

Where are the generated implementation plans saved?▼

Plans are written to the _scratchpad/plans/ directory with a timestamped filename like 2025-01-15-1430-add-auth.md. Each plan includes a YAML frontmatter block tracking its origin ticket, research documents, and related plans.

Does the planning process work with different AI coding agents?▼

Yes, the instructions are written as behavior contracts rather than provider-specific syntax, so they work across OpenCode, Codex, and Claude. If a named tool is unavailable, the closest equivalent behavior is used.

When is external research required during planning?▼

External research is mandatory when the plan touches security, authentication, billing, data migrations, privacy, compliance, or public API contracts. For localized refactors or low-risk internal changes it is optional, and the decision is documented either way.