Model Context Protocol moderate risk

Any Chat Completions

Integrates any OpenAI SDK compatible chat completion API (OpenAI, Perplexity, Groq, etc.) with Claude via the Model Context Protocol.

Connections & Capabilities

Connects To

GitHub

Capabilities

readwrite

Quickstart

Install

npm install
```

Config

{
  "mcpServers": {
    "chat-openai": {
      "command": "npx",
      "args": [
        "@pyroprompts/any-chat-completions-mcp"
      ],
      "env": {
        "AI_CHAT_KEY": "OPENAI_KEY",
        "AI_CHAT_NAME": "OpenAI",
        "AI_CHAT_MODEL": "gpt-4o",
        "AI_CHAT_BASE_URL": "https://api.openai.com/v1"
      }
    }
  }
}

Exposed MCP Tools (1)

moderate

chat

Relays a user's question to a configured AI chat provider and returns the response.

Relays user input to external service, potential for data leakage or prompt injection.

Safety Assessment

This server is relatively safe when used with trusted AI providers and when API keys are securely managed. However, it becomes risky if used with untrusted providers or if environment variables are not properly protected, potentially leading to data exposure or prompt injection vulnerabilities.

Relies on environment variables for API keys, reducing hardcoding.
Limited to chat completions, restricting broader system access.
Uses the Model Context Protocol, providing a defined communication interface.
Can be configured with different AI providers, allowing for risk diversification.

API keys stored as environment variables can be exposed if not properly managed.
The 'chat' tool relays user questions to external AI providers, potentially exposing sensitive information.
No built-in input sanitization, relying on the external AI provider for security.
Potential for prompt injection attacks if the AI provider is vulnerable.