Agent Plugins Marketplace
← All plugins

delegating-to-external-llm

v2.8.2

Claude is the architect: it writes specs and verdicts and picks a reasoning effort per task. The relay model (follows your codex config) on an OpenAI-compatible relay does all the implementation — agentically through the Codex CLI by default, or as chat completions applied verbatim as the fallback — and a clean-context Claude advisor reviews every deliverable before it ships.

Claude Code1 Skill

By YPYLicense: MIT3 GitHub starsUpdated 2 days ago

Directory evidence

Runtimes
Claude Code
Parsed components
1 skill or MCP entry
Source updated
Sep 21, 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 delegating-to-external-llm 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 delegating-to-external-llm@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/ypyik0669/delegating-to-external-llm

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

Plugin files

delegating-to-external-llm/
├── .claude-plugin/plugin.json
└── skills/delegating-to-external-llm/SKILL.md

Included Skills1

delegating-to-external-llmskills/delegating-to-external-llm/SKILL.md

Use when the user wants coding work done by an external model instead of by you — "delegate all coding", "you are only the brain", "only orchestrate", "don't write code yourself", "use GPT / codex / the relay to do it", "让外部模型写", "用 GPT 做" — or when the usage ledger shows the external model was barely used. Routes analysis, implementation and cross-vendor review to the relay model through deterministic scripts, so Claude spends tokens only on decisions, a spec review, and a verdict.

Plugin manifests1

.claude-plugin/plugin.json
{
  "name": "delegating-to-external-llm",
  "version": "2.8.2",
  "description": "Claude is the architect: it writes specs and verdicts and picks a reasoning effort per task. The relay model (follows your codex config) on an OpenAI-compatible relay does all the implementation — agentically through the Codex CLI by default, or as chat completions applied verbatim as the fallback — and a clean-context Claude advisor reviews every deliverable before it ships.",
  "author": {
    "name": "YPY"
  },
  "license": "MIT",
  "keywords": [
    "agents",
    "model-routing",
    "orchestration",
    "architect",
    "codex",
    "relay",
    "reasoning-effort",
    "delegation"
  ]
}

For maintainers

If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.

[delegating-to-external-llm on Agent Plugins Marketplace](https://pluginsmp.com/plugins/delegating-to-external-llm)