Does this skill work for small bug fixes?

Yes. The skill uses complexity-based planning. For simple tasks, it uses fewer agents and rounds to generate a concise plan, whereas major architectural changes trigger a more robust planning process.

How does the multi-agent discussion work?

The skill selects 3 to 10 specialized AI agents based on the task's complexity. These agents engage in 2 to 3 rounds of discussion to identify edge cases, refine architecture, and reach a consensus on the best implementation path.

Where are the implementation plans stored?

All generated plans are saved as structured markdown files in the .claude-output/ directory, typically following a 'plan-[timestamp].md' naming convention.

Can I review the plan before applying changes?

Absolutely. This skill is specifically designed as the planning step in a four-part workflow: Explore, Plan, Review, and Execute. You can review the output file manually or use the /review-plan command.

Implementation Planner

Name: Implementation Planner
Author: Uniswap

byUniswap

•

Productivity & Workflow

Generates detailed, multi-agent refined implementation plans for complex software features and architectural changes.

About

The Implementation Planner skill automates the critical design phase of software development by orchestrating a collaborative discussion between multiple specialized AI agents. By analyzing your existing codebase and project context, it synthesizes high-level feature requests into structured, actionable blueprints that include API designs, file change summaries, and step-by-step execution paths. This skill is essential for tackling complex refactors, migrations, or new feature additions where architectural consistency and risk mitigation are paramount.

Key Features

Multi-agent collaborative refinement for thorough architectural vetting
Automatic context synthesis from prior codebase exploration
Complexity-based planning tiers ranging from simple fixes to major migrations
13 GitHub stars
Structured markdown output including scope, API design, and sequential steps
Seamless integration with review and execution workflow commands

Use Cases

Planning a large-scale migration of a legacy database layer to a new API
Refactoring complex frontend logic to implement a unified design system
Architecting a secure user authentication system with multi-factor support

About

Key Features

Multi-agent collaborative refinement for thorough architectural vetting
Automatic context synthesis from prior codebase exploration
Complexity-based planning tiers ranging from simple fixes to major migrations
13 GitHub stars
Structured markdown output including scope, API design, and sequential steps
Seamless integration with review and execution workflow commands

Use Cases

Planning a large-scale migration of a legacy database layer to a new API
Refactoring complex frontend logic to implement a unified design system
Architecting a secure user authentication system with multi-factor support