claude-routing

Route tasks to Claude subagents based on evaluated complexity.

Updated Mar 9, 2026
One-click install
npx skills add https://github.com/cbbkrd-tech/jl-finishes --skill claude-routing
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-routing
Source: https://github.com/cbbkrd-tech/jl-finishes/tree/main/.claude/skills/claude-routing
Command: npx skills add https://github.com/cbbkrd-tech/jl-finishes --skill claude-routing

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured way to route tasks to specific Claude subagents (Haiku, Sonnet, Opus) based on complexity, ensuring efficient resource utilization and task management.

Core Features & Use Cases

  • Explicit Routing: Manually direct tasks to the most appropriate Claude model tier.
  • Subagent Management: Track and resume persistent subagents within a session.
  • Use Case: When faced with a complex debugging task, you can use /route zdebuguj memory leak w worker thread to ensure it's handled by the powerful Opus model, while simpler tasks like finding code snippets are routed to Haiku.

Quick Start

Use the claude-routing skill to route the task 'implement OAuth2 login' to the Sonnet model.

Frequently Asked Questions about claude-routing

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

FAQPage Schema
How do I route tasks to different Claude models based on complexity?

You can route tasks to different Claude models by evaluating their complexity and delegating them to specific subagents like Haiku for simple code generation and Opus for complex debugging.

Can I resume a persistent Claude subagent within an active session?

Yes, you can resume persistent subagents within a session by scanning conversation history for existing agent IDs to continue task delegation without reinitializing.

How do I delegate a complex debugging task to the Opus subagent?

You can delegate a complex debugging task to the Opus subagent by using explicit routing commands to ensure resource-intensive problems are handled by the highest complexity model tier.

What is the best way to manage AI subagents for tasks of varying difficulty?

Managing AI subagents by routing tasks based on evaluated complexity ensures efficient resource utilization, matching simple code snippets to Haiku and advanced logic to Sonnet.

When should I use explicit routing instead of default task management?

You should use explicit routing when you need to manually direct tasks to specific Claude model tiers to optimize resource utilization and verify the status of persistent subagents.

Does the Claude routing skill work without external dependencies?

Yes, the Claude routing skill operates without external dependencies, managing subagent initialization and status verification entirely within your session.