MCP server CX Codex

GoModel — Claude Code MCP server: features and setup

Q: How do I install GoModel for Claude Code?

On macOS or Linux, run `curl -fsSL https://gomodel.enterpilot.io/install.sh | sh`, then `gomodel`. Open `http://localhost:8080/admin/dashboard` and add a provider.

Q: Which API formats does GoModel accept?

OpenAI-compatible requests use `/v1`; Anthropic-compatible requests use `/v1/messages`. Official SDKs can connect by setting their base URL.

GoModel is a self-hosted gateway for model requests and MCP servers. It accepts OpenAI- and Anthropic-compatible API formats.

Open on GitHub

What GoModel does

GoModel accepts requests through the OpenAI-compatible /v1 API and Anthropic-compatible /v1/messages API. Official SDKs can use it by changing their base URL. Add a provider in the dashboard, then configure the gateway with .env, config.yaml, or the dashboard itself. The Docker setup maps port 8080 and stores data in the gomodel_data volume.

Features listed include response caching, cost tracking, budgets, and request limits. Virtual models can route requests between targets under stable model names. A provider must be configured before the sample request to /v1/responses can work. macOS and Linux also have an install script.

What it helps with

route model requests through one API
set spending limits for users, teams, or keys
cache repeated model responses
route requests between targets with virtual model names

Example prompts for the agent

Configure GoModel as the OpenAI SDK base URL and send a model request.
Add a provider in the GoModel dashboard, then test /v1/responses.
Set a team budget and show its spending in the dashboard.

How to install GoModel

Install

The MCP registry has no launch command for it: see the repository README

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about GoModel

How do I install GoModel for Claude Code?

On macOS or Linux, run curl -fsSL https://gomodel.enterpilot.io/install.sh | sh, then gomodel. Open http://localhost:8080/admin/dashboard and add a provider.

What does GoModel do?

GoModel routes model requests through OpenAI- and Anthropic-compatible APIs. It also offers caching, cost tracking, and budget controls.

Which API formats does GoModel accept?

OpenAI-compatible requests use /v1; Anthropic-compatible requests use /v1/messages. Official SDKs can connect by setting their base URL.

Similar MCP servers

All MCP servers for Agent setup: memory, context, orchestration →

supabase-schema-architect davila7 A subagent that designs PostgreSQL schemas in Supabase, plans migrations and writes RLS policies. Useful when your database grows and access rules need sorting out. 3,839installs fetcher fetcher-sh Shared payment and setup skill for fetcher.sh: API keys, credits, x402 micropayments and MCP server config. Meant to run before or alongside the per-service skills. CX Codex +1 3,700installs tigris-object-operations tigrisdata A skill that lets your agent upload, download, delete and list objects in Tigris Storage, and mint presigned URLs. CX Codex +1 3,200installs building-tools celigo A skill for building reusable Celigo tool resources: lookups, imports, transforms and branching behind one input/output contract. CX Codex +1 2,900installs dm-limits-and-best-practices cognitedata A reference skill on CDF Data Modeling limits: concurrency and 429s, pagination, batching, QueuedTaskRunner. Useful when your code starts failing under load. CX Codex +1 1,900installs prismic prismicio A skill for working with Prismic: it manages content models, repository settings and docs through the CLI. Useful when your project runs on Prismic. CX Codex +1 1,500installs