Engineering
implementation-planner avatar

implementation-planner

Create structured, high-quality technical implementation plans via an agent-driven, iterative process. Ideal for complex refactoring, new features, and technical design.

Introduction

The implementation-planner is a specialized skill designed to streamline technical planning by orchestrating an agent-first, interactive workflow. It is intended for software engineers, architects, and technical leads who need to navigate complex development tasks, such as significant architectural refactors, feature implementation, or system upgrades. By delegating data gathering and verification to parallel subagents, the planner ensures that the final documentation is based on comprehensive, verified research.

  • Automated GitHub issue analysis via dedicated subagents, covering descriptions, comments, and related pull requests.

  • Parallel research capabilities using Task agents to explore codebases, analyze design patterns, and verify testing strategies simultaneously.

  • Language-agnostic foundation that integrates seamlessly with language-specific guideline skills (e.g., go-dev-guidelines) to ensure architectural and code standards compliance.

  • Structured file generation, creating clear documentation for implementation tasks, success criteria, and testing strategies.

  • Iterative refinement process that separates working notes from finalized deliverables for cleaner project history.

  • Use this skill at the beginning of any significant coding project to establish a clear, documented path forward.

  • Always identify the project's primary language (Go, Python, TypeScript, etc.) immediately to trigger the appropriate guidelines.

  • Input requirements include either a specific GitHub issue number for context-tracking or a project-specific name for ad-hoc planning.

  • Expected outputs include a complete plan directory structure containing design documents, todo lists, and verification strategies.

  • Practitioners should remain active in the 'human-in-the-loop' phase, using the generated findings to make informed design decisions before the agent finalizes the implementation files.

  • The planner heavily utilizes the Task tool for background research; ensure sufficient permissions for repository navigation and issue retrieval.

Repository Stats

Stars
261
Forks
29
Open Issues
32
Language
Go
Default Branch
main
Sync Status
Idle
Last Synced
May 1, 2026, 08:35 AM
View on GitHub