Skill

delegate-to-ai

Route tasks to external AI models via PAL MCP tools

Install

npx claudepluginhub jacobpevans/claude-code-plugins --plugin ai-delegation

Tool Access

This skill uses the workspace's default tool permissions.

Preview

Routes tasks to specialized models based on task type and constraints using PAL MCP tools.

SKILL.md

Similar Skills

cache-components

Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.

cache-components

139.2k

mcp-builder

9 files

Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).

anthropics-skills-13

124.2k

canvas-design

20 files

Generates original PNG/PDF visual art via design philosophy manifestos for posters, graphics, and static designs on user request.

anthropics-skills-13

124.2k

Stats

Parent Repo Stars2

Parent Repo Forks0

Last CommitMar 22, 2026

Actions

View Source View Plugin View on GitHub View README

Delegate to External AI

Routes tasks to specialized models based on task type and constraints using PAL MCP tools.

When to Delegate

Delegate when Claude is not the best tool:

Large context (1M+ tokens) -> Gemini 3 Pro via PAL chat tool

Math/reasoning -> DeepSeek R1 via PAL chat tool

Private/offline -> Local Ollama via PAL chat tool

Code review consensus -> Multi-model via PAL consensus tool

Architecture planning -> Claude Opus or Gemini via PAL planner tool

PAL MCP Tools

chat - Single model for straightforward tasks

clink - Parallel queries across multiple models

consensus - Multi-model agreement for critical decisions

codereview - Structured code review with multiple perspectives

planner - Architecture and design planning

precommit - Quick validation before committing

Workflow

Identify task type (research, coding, review, architecture, planning)

Select PAL MCP tool based on type and constraints

Execute via Task tool with appropriate agent subtype

Synthesize results if using multi-model tools

Model Routing

Task Type

Cloud Model

Local Model

PAL Tool

Research

Gemini 3 Pro

qwen3-next:80b

chat, clink

Complex Coding

Claude Opus

qwen3-coder:30b

codereview

Fast Tasks

Claude Sonnet

qwen3-next:latest

chat

Code Review

Multi-model

deepseek-r1:70b

consensus

Architecture

Claude Opus

qwen3-next:80b

planner

Local-Only Mode

When localOnlyMode is enabled or --local flag is passed, all tasks route to Ollama models. No cloud API calls are made.

Related Skills

auto-maintain (ai-delegation)

Delegate to External AI

Routes tasks to specialized models based on task type and constraints using PAL MCP tools.

When to Delegate

Delegate when Claude is not the best tool:

Large context (1M+ tokens) -> Gemini 3 Pro via PAL chat tool
Math/reasoning -> DeepSeek R1 via PAL chat tool
Private/offline -> Local Ollama via PAL chat tool
Code review consensus -> Multi-model via PAL consensus tool
Architecture planning -> Claude Opus or Gemini via PAL planner tool

PAL MCP Tools

chat - Single model for straightforward tasks
clink - Parallel queries across multiple models
consensus - Multi-model agreement for critical decisions
codereview - Structured code review with multiple perspectives
planner - Architecture and design planning
precommit - Quick validation before committing

Workflow

Identify task type (research, coding, review, architecture, planning)
Select PAL MCP tool based on type and constraints
Execute via Task tool with appropriate agent subtype
Synthesize results if using multi-model tools

Model Routing

Task Type	Cloud Model	Local Model	PAL Tool
Research	Gemini 3 Pro	qwen3-next:80b	`chat`, `clink`
Complex Coding	Claude Opus	qwen3-coder:30b	`codereview`
Fast Tasks	Claude Sonnet	qwen3-next:latest	`chat`
Code Review	Multi-model	deepseek-r1:70b	`consensus`
Architecture	Claude Opus	qwen3-next:80b	`planner`

Local-Only Mode

When localOnlyMode is enabled or --local flag is passed, all tasks route to Ollama models. No cloud API calls are made.

Related Skills

auto-maintain (ai-delegation)