"The most widely used models" matching MCP servers:

Lets MCP clients delegate tasks to the Google Antigravity CLI, including starting and continuing conversations and listing models, while returning structured results over stdio.

Delegates coding tasks to the locally installed Kimi Code CLI, reusing its auth, models, and configuration while adding tools for sessions, replies, model discovery, live turn steering, and history access.

Delegate tasks from Claude Code to other models (Codex CLI, DeepSeek, OpenRouter, etc.) without leaving the app.

Enables MCP clients to invoke Cursor SDK's agent runtime, run coding agents, list models, and continue conversations.

Provides free, unlimited access to Claude AI models via Puter.js for AI coding agents and IDEs through the MCP protocol.

Enables an orchestrator like Claude Code to hand coding work off to Cursor SDK agents, run as detached background jobs that survive the session and can be listed, inspected, steered, resumed, or stopped. Agents are asynchronous and reusable across follow-up turns, letting expensive frontier models delegate cheaply without blocking or paying to read the results.

Enables MCP clients to run Cursor CLI agent prompts with any model the installed CLI offers, continue existing sessions, and list available models and sessions. Adds guarded execution with concurrency limits, timeouts, cancellation, and operator-gated auto-approval so multiple models can be run in parallel.

Enables AI assistants to consult other AI models via OpenRouter for help with coding tasks.

MCP server for Python build123d to help AIs develop and reason about 3D models and CAD

Add OpenAI (Codex) models to Claude code

A lightweight MCP server that bridges OpenAI Codex to Gemini models via the Antigravity CLI, enabling natural language prompts and model listing within Codex.

Extended MCP server for OpenAI with support for GPT-5 family, Codex models, and automatic API endpoint selection, enabling chat interactions via the openai_chat tool.

Production-grade, autonomous Model Context Protocol (MCP) server that elevates AI models from stateless code generators into persistent, self-verifying software engineers.

Enables bounded, local coding tasks to be delegated to OpenAI-compatible models with explicit working directory and read/write paths, pre-application unified-diff validation, and optional verification commands. It routes tasks by complexity and risk, while blocking sensitive changes such as auth, payment, secrets, migrations, database, deployment, or production environments.

Enables coding agents to review GitHub pull requests with a panel of AI models through local or remote MCP endpoints.

Exposes the Quebragalho pre-paid model gateway to MCP clients, letting coding agents like Claude Code, Codex, Cursor and OpenCode spawn separate repo-aware sub-agents that run on 17 frontier models, with background job tracking, routing tiers and persistent per-project memory.

Enables MCP clients such as Claude Code to use the latest Codex CLI as a tool, starting new Codex threads and continuing them with replies, including approvals handling and progress notifications. It keeps the classic codex and codex-reply tools working while letting users swap in newer Codex binaries to unlock the newest models.

Enables coding agents to check human-stated assumptions against authoritative sources like AWS AppConfig, Sentry, and the repository, detecting when beliefs have drifted from reality. It provides tools to inspect belief history, list verifiable properties, review open drifts, and record verified understanding.

Launches the official Codex CLI as a persistent MCP server so DeepSeek Harness can invoke Codex models without storing or transmitting ChatGPT credentials to third parties.

Enables Claude to hand off implementation tasks to OpenAI Codex models running through the Codex CLI, with tools for listing models, delegating work synchronously or in the background, collecting results, and continuing past sessions, while Claude stays the orchestrator that reviews diffs and runs tests.