model-router

v1.1.0

Keep the expensive lead model on planning and review; route routine work to pinned cheaper workers (Haiku, Sonnet, Opus) and audit which models actually ran.

Claude Code2 Skills

By dd106-lang0 GitHub starsUpdated 5 days ago

Directory evidence

Runtimes: Claude Code
Parsed components: 2 skill or MCP entries
Source updated: Sep 18, 2026
Manifest status: Canonical path parsed

The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →

Install model-router for Claude Code

Installs for the current user

claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install model-router@agent-plugin-marketplace

Paste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.

The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.

Get the source manually

$ git clone https://github.com/dd106-lang/claude-model-router

Clone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/model-router/.

Plugin files

plugins/model-router/

├── .claude-plugin/plugin.json
├── skills/route/SKILL.md
└── skills/usage-audit/SKILL.md

Included Skills2

routeskills/route/SKILL.md

Multi-model delegation playbook. Protects the lead model's tokens (Fable or Opus) by keeping it on planning and review and pushing the work down to cheaper pinned workers - scout and runner on Haiku, builder on Sonnet, specialist on Opus. Use when the user asks to save tokens or usage, to delegate, or to use subagents or cheaper models, and when starting a multi-step coding, file-generation, or research task that contains routine work.

usage-auditskills/usage-audit/SKILL.md

Shows which Claude models actually ran and how many tokens each used, split into the lead and its subagents, for the current session or a time window. Reads the local Claude Code transcripts. Use to check that delegation to cheaper models is really happening, or when the user asks where their tokens or usage went by model.

Plugin manifests1

plugins/model-router/.claude-plugin/plugin.json

{
  "$schema": "https://anthropic.com/claude-code/plugin.schema.json",
  "name": "model-router",
  "version": "1.1.0",
  "description": "Keep the expensive lead model on planning and review; route routine work to pinned cheaper workers (Haiku, Sonnet, Opus) and audit which models actually ran.",
  "author": {
    "name": "dd106-lang"
  },
  "keywords": [
    "subagents",
    "model-routing",
    "cost",
    "tokens",
    "delegation"
  ]
}