# Chat Roulette

> Use this tool when you need to facilitate real-time chat conversations between users in a unique, interactive environment. It solves problems of creating engaging, dynamic communication experiences and testing MCP client capabilities, with inputs of user messages and outputs of real-time conversations. It is ideal for use cases such as interactive chat experiences, protocol testing, and demonstrating innovative applications of the Model Context Protocol.

Canonical page: https://skillsregistry.net/skills/chat-roulette  
JSON: https://api.skillsregistry.net/v1/skills/chat-roulette

## Description

This MCP server enables real-time chat conversations between users through an innovative abuse of the Model Context Protocol, using tool calls and long-polling to create a chat roulette-style experience. Built using Python with FastMCP, it implements a room-based system where users can join specific chat rooms or be randomly paired, with the `wait_for_message` tool blocking until messages arrive to simulate real-time communication without requiring custom notifications support from MCP clients. The implementation uses in-memory storage for simplicity, supports two-user rooms with message queuing, and works with any MCP client including Claude Code and standard HTTP tools, making it valuable for creating interactive chat experiences, testing MCP client capabilities, and demonstrating creative applications of the protocol beyond traditional tool-calling patterns.

## Trust

- **Trust score (0–1):** 0.92
- **Verification tier:** verified
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** other
- **Updated:** 2026-09-28

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/chat-roulette)
- **Repository:** <https://github.com/hbd/mcp-chat>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "chat-roulette"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/chat-roulette` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/chat-roulette/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt