multi-llm-mcp-728
Active·★ 56·MIT·Updated 2026-06-04
This project is an MCP tool for Claude Code, designed to facilitate task execution via Codex CLI and enable parallel calling of various large language models such as GPT, Kimi, DeepSeek, and Qwen. It provides a unified framework for orchestrating diverse AI models and code execution environments, enhancing efficiency for complex developer workflows.
#Multi-LLM Orchestration#CodeX CLI Integration#Parallel AI#Agent Framework#Session Management#API Proxy#Developer Tooling#Workflow Automation
01
Features
01Supports parallel invocation of multiple large language models (e.g., GPT, Kimi, DeepSeek, Qwen, Claude) for tasks like analysis or comparison.
02Integrates with Codex CLI, allowing tasks to be dispatched and executed directly within a sandboxed code environment.
03Provides session management capabilities to clear or list active multi-model or Codex CLI interaction sessions.
04Offers a unified interface for interacting with various LLM providers, simplifying model configuration and usage.
02
Compatibility
Python
Runtime
Verified via docs
Claude Code
Integration
Verified via docs
Codex CLI
Tool
Verified via docs
OpenAI API
LLM Provider
Verified via docs
Moonshot API
LLM Provider
Verified via docs
DeepSeek API
LLM Provider
Verified via docs
03
Quick start
1
$ git clone https://github.com/Breezepaunveil/multi-llm-mcp-728.git
2
$ cd multi-llm-mcp-728
3
$ python setup.py
4
$ pip install fastmcp openai
04
Use cases
↳Simultaneously analyze code or proposed solutions using multiple LLMs to get diverse perspectives and identify potential issues.
↳Automate coding tasks like checking project structure, creating test files, or fixing code directly within the Claude Code environment via Codex CLI.
↳Dispatch specific tasks to different LLMs based on cost or desired context, for example, using a cheaper model for "grunt work" or an isolated Claude for unbiased judgment.
↳Engage in multi-turn conversational interactions with individual LLMs for in-depth queries or focused discussions.
05
Alternatives
polter★ 20
One Claude Code session supervises the agents in your other terminal tabs: which is stuck, which is done, which is still running. A fork of Ghostty.
agent-manager★ 625
The fastest workflow for every AI coding agent. Live status, quick prompts, worktrees, and diff review from one tmux TUI.
AI-company★ 375
Multi-agent team operating system for Claude Code. 108 MCP tools, 40+ agent templates, 10 lifecycle hooks, 7 pipeline workflows. Persistent teams, structured meetings, task wall, real-time React dashboard. No LangChain/AutoGen — pure CC native integration.
unterm★ 15
The terminal AI agents can drive — and the cockpit to run them from. 99 MCP methods / 21 namespaces, agent Inbox, verified worktree fleets, diff review & rollback. macOS / Linux / Windows, local-first, $0, MIT.
entwurf★ 31
Garden-citizen dispatch for pi, Claude Code, Copilot CLI, OMP, Antigravity, and Codex; visible tmux sibling seats for Pi/Claude/Copilot/OMP/Codex, preserving each harness's auth, tools, and transcript; plus Claude/Cortex ACP.
controlkeel★ 10
Agent control plane for governed AI coding: validate changes, enforce policy gates, track findings, proofs, and evals based on your habits.
Comments
Log in to leave a comment
No comments yet. Be the first!