MehrdadNsr

Multi AI Delegation — AI skill for Claude Code

AI community

A Claude skill (plugin) that splits big jobs across your other AI chat sites, such as ChatGPT, Grok, Gemini, DeepSeek, Kimi, Qwen and Perplexity, through Claude in Chrome.

How to install Multi AI Delegation

This entry records only its repository, not the path inside it, so there is no exact command to give. Open MehrdadNsr/multi-ai-delegation and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Multi AI Delegation does

A Claude skill (plugin) that splits big jobs across your other AI chat sites, such as ChatGPT, Grok, Gemini, DeepSeek, Kimi, Qwen and Perplexity, through Claude in Chrome. You approve a plan with sites, cost and time first; Claude then pilots, runs, checks, merges and reports. Free-only setups work. Beta.

Alternatives in AI

  • Omniget — Udemy & Hotmart course downloader, YouTube downloader (yt-dlp GUI, 1,800+ sites) + desktop app for AI agents 13.9k ★
  • AGENTS.md - Skill Seekers — Concise reference for AI coding agents 11.1k ★
  • EchoBird — One-click install + model switch:Claude Code,Codex CLI (OpenAI), Grok Build (xAI), DeepSeek Harness, Kimi Code 3.1k ★

README

multi-ai-delegation

A Claude skill, packaged as a plugin, that lets Claude split big jobs across your other AI chat sites and tools: ChatGPT, Grok, Gemini, Google AI Studio, Gemini Notebook (NotebookLM), Perplexity, DeepSeek, Kimi, Qwen and OpenRouter, plus Suno, ElevenLabs and Higgsfield for media. Claude drives the sites in your own browser through the Claude in Chrome extension. Nothing is sent until you approve a plan that shows which site gets what, at what cost and how long it will take.

**Status: beta.** Tested on one account in October 2026; see [Status](#status).

What it does

  • Splits big material (chat exports, logs, transcripts, long PDFs) into chunks with stable ids and routes each chunk by cost tier: Claude itself, Claude subagents, your subscriptions, free sites, and pay-per-use tools only when nothing cheaper can do the job.
  • Shows a plan table first (site, tier, mode, chunks, volume, prompts, quota share or cost, time) and waits for your go.
  • Pilots one small chunk per worker, runs the rest in parallel, checks every answer mechanically, spot-checks it against the source, plants known-answer canaries, and cross-checks high-stakes work on a second model family.
  • Merges the results and reports who did what, what was verified and what was not, what was spent, and which chats were left in each site's history.
  • Also: a council (for, against and neutral views from different model families), a review chain and challenge pass, media jobs with a cost quote first, and briefs for a coding agent you drive.

Example:

**You:** Here's a chat export with about 1,200 messages. Pull out every decision with its date and owner. Split it across my AIs.

**Claude:** *(plan table)* Claude subagents take L0001–L0400, ChatGPT L0401–L0800, Grok L0801–L1200, one pilot chunk each; 16 prompts, about 20 minutes; fallback DeepSeek; personal data: none. Go?

**You:** go

**Claude:** runs the pilots, checks them, sends the rest, spot-checks and merges one CSV