How does it ensure code quality across different agents?

The orchestrator grounds itself in your project's CLAUDE.md, design principles, and existing code patterns before crafting highly specific prompts for each sub-agent.

Can it handle dependencies between implementation tasks?

Yes, the orchestrator can plan sequential workflows where one agent waits for another's output, or parallel workflows for independent components.

When should I use the Implementation Orchestrator skill?

This skill is best for large designs involving 20+ files or projects with independent subsystems that can be implemented in parallel by different agents.

Which models are used by this skill?

It utilizes Claude Opus for high-level reasoning and planning, and spawns Claude Sonnet agents to handle the granular implementation tasks.

Implementation Orchestrator

Name: Implementation Orchestrator
Author: nklisch

bynklisch

•

Productivity & Workflow

Orchestrates complex code implementations by coordinating multiple AI agents to execute large-scale design documents.

The Implementation Orchestrator skill leverages the high-reasoning capabilities of Claude Opus to manage large-scale development tasks. It serves as a project manager for your codebase, reading complex design documents, analyzing existing project patterns, and delegating specific coding tasks to focused Sonnet agents. By grounding implementation in your project's specific architectural principles and CLAUDE.md conventions, it ensures that even massive changes involving 20+ files remain consistent, high-quality, and properly integrated.

Key Features

01Automated work splitting for large design documents into agent-sized units

02Deep grounding in project-specific architectural and implementation principles

03Support for both parallel and sequential task execution workflows

04Automated end-to-end verification and testing of all implemented code

052 GitHub stars

06High-precision prompt generation for specialized sub-agent coordination

Use Cases

01Building complex subsystems in parallel using multiple specialized task agents

02Implementing large-scale features or refactors spanning 20 or more files

03Enforcing strict adherence to architectural patterns across a massive codebase

Key Features

01Automated work splitting for large design documents into agent-sized units

02Deep grounding in project-specific architectural and implementation principles

03Support for both parallel and sequential task execution workflows

04Automated end-to-end verification and testing of all implemented code

052 GitHub stars

06High-precision prompt generation for specialized sub-agent coordination

Use Cases

01Building complex subsystems in parallel using multiple specialized task agents

02Implementing large-scale features or refactors spanning 20 or more files

03Enforcing strict adherence to architectural patterns across a massive codebase