What dependencies does MCP OpenAI Server need?
Node.js 18 or higher and an OpenAI API key are required. The server also depends on the Claude Desktop app for runtime configuration.
@mzxrai
基本信息
Chat with OpenAI models from Claude Desktop
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"mcp-openai": {
"command": "npx",
"args": [
"-y",
"@mzxrai/mcp-openai@latest"
],
"env": {
"OPENAI_API_KEY": "your-api-key-here (get one from https://platform.openai.com/api-keys)"
}
}
}
}Use this tool when a user specifically requests to use one of OpenAI's models (gpt-4o, gpt-4o-mini, o1-preview, o1-mini). This tool sends messages to OpenAI's chat completion API using the specified model.
MCP OpenAI Server is a Model Context Protocol (MCP) server that integrates OpenAI’s chat models directly into Claude. It is built for users of Claude Desktop who want to tap into OpenAI’s models (gpt-4o, gpt-4o-mini, o1-preview, o1-mini) from within the same interface.
Install the Claude Desktop app, obtain an OpenAI API key, then add the provided JSON entry to your claude_desktop_config.json file. Once configured, simply ask Claude to query an OpenAI model—for example, “Can you ask o1 what it thinks about this problem?”—and the server handles the rest.
openai_chat) with required messages and optional modelNode.js 18 or higher and an OpenAI API key are required. The server also depends on the Claude Desktop app for runtime configuration.
On macOS, the JSON configuration lives at `~/Library/Application\ Support/Claude/claude_desktop_config.json`.
macOS is verified; Linux is not yet supported.
It exposes one tool, `openai_chat`, which accepts a required `messages` array and an optional `model` parameter (defaults to gpt-4o).
The server expects your OpenAI API key to be set in the environment variable `OPENAI_API_KEY` inside the MCP server configuration.
Detect and redact PII, PHI, PCI and secrets — SSNs, credit cards, passports, API keys and cloud credentials — in text and in local files, using the Strac DLP API. Five tools: redact_text, detect_sensitive_data, detect_fi

ATL doesn’t just help agents find tools — it gets the job done. It finds and chooses the right AI provider, tool, or MCP server, executes the task, and switches to a fallback when a provider fails. It can optimize for lo
GoBuy gives AI agents marketplace-evidence trust scores. One call returns a 0–100 Evidence Score with observed/missing signals, freshness, and product context. Tools: • check_product_trust — inspect one Amazon, Walmart,
MCP-native LLM councils for debates, juries, blind panels, voting, refinement, and custom multi-model deliberation.
Open-source Node.js FHIR MCP server with SMART Backend Services, metadata-aware search/CRUD tools, compact responses, FHIRPath filtering, safe pagination, audit events, and terminology lookup.
评论