OpenNews MCP

OpenNews MCP

A cryptocurrency news aggregation server providing AI-powered sentiment ratings, trading signals, and real-time market updates. It allows users to search news by coin, source, or engine type and filter results based on AI impact scores and bullish/bearish signals.

Category
访问服务器

README

<p align="center"> <b>OpenNews MCP Server</b><br> Crypto News Aggregation · AI Ratings · Trading Signals · Real-time Updates </p>

<p align="center"> <a href="./README.md">English</a> | <a href="./docs/README_ZH.md">中文</a> | <a href="./docs/README_JA.md">日本語</a> | <a href="./docs/README_KO.md">한국어</a> </p>


Quick Install

First, get your API Token at https://6551.io/mcp.

Claude Code

claude mcp add opennews \
  -e OPENNEWS_TOKEN=<your-token> \
  -- uv --directory /path/to/opennews-mcp run opennews-mcp

Replace /path/to/opennews-mcp with your local project path, and <your-token> with your API token.

OpenClaw

export OPENNEWS_TOKEN="<your-token>"
cp -r openclaw-skill/opennews ~/.openclaw/skills/

Let AI Review and Install

Not sure if this MCP is safe? Send the following prompt to your AI assistant to review the source code before installing:

Copy the prompt below and paste it to your AI assistant (replace <project-path> and <your-token> with actual values):

Please help me review and install the opennews-mcp MCP server. The project is at <project-path>.

Steps:
1. Review the security of the following files:
   - src/opennews_mcp/api_client.py — Confirm it only connects to ai.6551.io, no data sent elsewhere
   - src/opennews_mcp/config.py — Confirm token is only read from local config.json or env vars, not hardcoded or leaked
   - src/opennews_mcp/tools/*.py — Confirm all tools only do API queries, no file writes, command execution, or dangerous operations
   - pyproject.toml — Confirm dependencies are only mcp, httpx, websockets, no suspicious packages
2. Tell me your conclusion: Safe / Risky / Problematic, and the specific reasons
3. If safe, help me install (Token from https://6551.io/mcp):
   claude mcp add opennews -e OPENNEWS_TOKEN=<your-token> -- uv --directory <project-path> run opennews-mcp

What Can It Do?

After connecting, just tell your AI assistant:

You Say It Does
"Latest crypto news" Get latest articles
"Search SEC regulation news" Full-text keyword search
"BTC related news" Filter by coin
"Bloomberg articles" Filter by source
"On-chain events" Filter by engine type (onchain)
"Important news with AI score above 80" High score filtering
"Bullish signals" Filter by trading signal (long)
"Subscribe to real-time news" WebSocket live updates

Available Tools

Category Tool Description
Discovery get_news_sources Get all news source category tree
list_news_types All available news source codes
Search get_latest_news Latest articles
search_news Keyword search
search_news_by_coin By coin (BTC, ETH, SOL...)
get_news_by_source By engine type and source
get_news_by_engine By type (news, listing, onchain, meme, market)
search_news_advanced Advanced search (multiple filters)
AI get_high_score_news Articles with score >= threshold
get_news_by_signal By signal: long / short / neutral
Real-time subscribe_latest_news WebSocket real-time collection

Configuration

Get API Token

Get your API Token at https://6551.io/mcp.

Set environment variable:

# macOS / Linux
export OPENNEWS_TOKEN="<your-token>"

# Windows PowerShell
$env:OPENNEWS_TOKEN = "<your-token>"
Variable Required Description
OPENNEWS_TOKEN Yes 6551 API Bearer Token (from https://6551.io/mcp)
OPENNEWS_API_BASE No Override REST API URL
OPENNEWS_WSS_URL No Override WebSocket URL
OPENNEWS_MAX_ROWS No Max results per request (default 100)

Also supports config.json in project root (env vars take precedence):

{
  "api_base_url": "https://ai.6551.io",
  "wss_url": "wss://ai.6551.io/open/news_wss",
  "api_token": "<your-token>",
  "max_rows": 100
}

Data Structure

Each article returns:

{
  "id": "unique-article-id",
  "text": "Title / Content",
  "newsType": "Bloomberg",
  "engineType": "news",
  "link": "https://...",
  "coins": [{ "symbol": "BTC", "market_type": "spot", "match": "title" }],
  "aiRating": {
    "score": 85,
    "grade": "A",
    "signal": "long",
    "status": "done",
    "summary": "Chinese summary",
    "enSummary": "English summary"
  },
  "ts": 1708473600000
}
AI Field Description
score 0-100 impact score
signal long (bullish) / short (bearish) / neutral
status done = AI analysis completed

<details> <summary><b>Manual Installation for Other Clients</b> (click to expand)</summary>

In all configurations below, replace /path/to/opennews-mcp with your actual local project path, and <your-token> with your token from https://6551.io/mcp.

Claude Desktop

Edit config file (macOS: ~/Library/Application Support/Claude/claude_desktop_config.json, Windows: %APPDATA%\Claude\claude_desktop_config.json):

{
  "mcpServers": {
    "opennews": {
      "command": "uv",
      "args": ["--directory", "/path/to/opennews-mcp", "run", "opennews-mcp"],
      "env": {
        "OPENNEWS_TOKEN": "<your-token>"
      }
    }
  }
}

Cursor

~/.cursor/mcp.json or Settings > MCP Servers:

{
  "mcpServers": {
    "opennews": {
      "command": "uv",
      "args": ["--directory", "/path/to/opennews-mcp", "run", "opennews-mcp"],
      "env": {
        "OPENNEWS_TOKEN": "<your-token>"
      }
    }
  }
}

Windsurf

~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "opennews": {
      "command": "uv",
      "args": ["--directory", "/path/to/opennews-mcp", "run", "opennews-mcp"],
      "env": {
        "OPENNEWS_TOKEN": "<your-token>"
      }
    }
  }
}

Cline

VS Code sidebar > Cline > MCP Servers > Configure, edit cline_mcp_settings.json:

{
  "mcpServers": {
    "opennews": {
      "command": "uv",
      "args": ["--directory", "/path/to/opennews-mcp", "run", "opennews-mcp"],
      "env": {
        "OPENNEWS_TOKEN": "<your-token>"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}

Continue.dev

~/.continue/config.yaml:

mcpServers:
  - name: opennews
    command: uv
    args:
      - --directory
      - /path/to/opennews-mcp
      - run
      - opennews-mcp
    env:
      OPENNEWS_TOKEN: <your-token>

Cherry Studio

Settings > MCP Servers > Add > Type stdio: Command uv, Args --directory /path/to/opennews-mcp run opennews-mcp, Env OPENNEWS_TOKEN.

Zed Editor

~/.config/zed/settings.json:

{
  "context_servers": {
    "opennews": {
      "command": {
        "path": "uv",
        "args": ["--directory", "/path/to/opennews-mcp", "run", "opennews-mcp"],
        "env": {
          "OPENNEWS_TOKEN": "<your-token>"
        }
      }
    }
  }
}

Any stdio MCP Client

OPENNEWS_TOKEN=<your-token> \
  uv --directory /path/to/opennews-mcp run opennews-mcp

</details>


Compatibility

Client Installation Status
Claude Code claude mcp add One-click
OpenClaw Copy Skill directory One-click
Claude Desktop JSON config Supported
Cursor JSON config Supported
Windsurf JSON config Supported
Cline JSON config Supported
Continue.dev YAML / JSON Supported
Cherry Studio GUI Supported
Zed JSON config Supported

Related Projects


Development

cd /path/to/opennews-mcp
uv sync
uv run opennews-mcp
# MCP Inspector test
npx @modelcontextprotocol/inspector uv --directory /path/to/opennews-mcp run opennews-mcp

Project Structure

├── README.md
├── openclaw-skill/opennews/   # OpenClaw Skill
├── knowledge/guide.md         # Embedded knowledge
├── pyproject.toml
├── config.json
└── src/opennews_mcp/
    ├── server.py              # Entry point
    ├── app.py                 # FastMCP instance
    ├── config.py              # Config loading
    ├── api_client.py          # HTTP + WebSocket
    └── tools/                 # Tools

License

MIT

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选