This project provides an MCP (Model Context Protocol) server for aggregating and analyzing cryptocurrency news. It targets AI assistants and developers seeking to integrate real-time, AI-enriched crypto market intelligence into their workflows, offering benefits like automated news filtering, sentiment analysis, and trading signal generation.

How It Works

The OpenNews MCP server acts as a bridge to the 6551.io API, fetching crypto news articles. It processes these articles using AI to generate impact scores, trading signals (bullish/bearish), and summaries. The system supports advanced querying, including filtering by cryptocurrency symbol, news source, and AI-derived metrics, with real-time updates delivered via WebSockets. This approach centralizes AI-driven crypto news analysis for seamless integration into AI agent environments.

Quick Start & Requirements

To install, first obtain an API token from https://6551.io/mcp. The primary installation method involves running the server using uv, a Python package runner. For example, using Claude Code: claude mcp add opennews -e OPENNEWS_TOKEN=<your-token> -- uv --directory /path/to/opennews-mcp run opennews-mcp. Alternatively, for OpenClaw, copy the skill directory: export OPENNEWS_TOKEN="<your-token>"; cp -r openclaw-skill/opennews ~/.openclaw/skills/. Development setup requires uv sync within the project directory. Key requirements include a valid OPENNEWS_TOKEN and the uv package manager.

**