# MCP Conversation Server A Model Context Protocol (MCP) server implementation for managing conversations with OpenRouter's language models. This
MCP Conversation Server manages OpenRouter conversations through MCP tools and resources. It stores conversation history in a configurable local directory and routes messages to the model IDs configured by the application.
create-conversation starts a conversation with an OpenRouter model and optional title.send-message adds a message to an existing conversation, with optional streaming.list-conversations lists conversations using the documented filters.The current README loads settings from config/models.yaml: the OpenRouter API key, persistence path, model configuration and default model. It explicitly states that environment variables are no longer required. A placeholder value for the API key is not an environment-variable name.
The project documents installing the package and constructing ConversationServer in application code. That is not evidence that an arbitrary npx command will launch a standalone server. Follow the documented application entry point and test it with your MCP client before relying on it.
Conversation resources expose message history and metadata. Choose the persistence directory deliberately and verify file permissions. Test recovery after a restart and how your chosen OpenRouter model handles streaming and context limits.
Browse AI service MCP integrations or hosted alternatives.
Check the linked publisher documentation and your connected client's tool list. Names and availability can vary with the server version and account permissions.
| Tool | What it does |
|---|---|
| create-conversation | Start an OpenRouter conversation with a configured model. |
| send-message | Send a message to a conversation, optionally streaming the response. |
| list-conversations | List conversations with supported filters. |
Node.js application integration, an OpenRouter API key and config/models.yaml settings.
The original Chromium automation reference server — simple, screenshot-driven browser control.
Build a programmable telecommunications stack for connecting telephony services with the Internet via a cloud-based utility.
Automates browsers and APIs, enabling Large Language Models (LLMs) to interact with web pages for tasks like screenshot capture and JavaScript
Workers, KV, R2 and D1 by conversation — Cloudflare's official remote servers for building and observability.
Search built for AI, not humans — semantic web search that returns model-ready content, plus code context.
Answers, not links — delegate questions to Perplexity's search-grounded models and get cited responses back.