TrendRadar

You can use TrendRadar as an MCP server to analyze and push news data with AI-assisted insights. This MCP enables you to run a local or remote analysis service that exposes an MCP API, so clients can query, analyze, translate, and compare news data gathered by TrendRadar. It supports HTTP-based MCP access and also local, STDIO-style execution for tight integration with editors and tools. This guide walks you through practical usage, installation steps, and key configuration notes so you can start querying and analyzing news efficiently.

How to use

Connect to the MCP server using either the HTTP URL or a local STDIO setup, then interact with supported MCP features to retrieve, analyze, and compare trending news data. Practical use cases include querying latest topics, performing trend analysis over time, fetching RSS sources, translating or translating-and-analyzing content, and using AI-powered analysis modules to generate insights and summaries.

How to install

Prerequisites you need before installation are Docker and/or a local runtime for the STDIO mode. If you plan to deploy with Docker, you will pull and run the TrendRadar images. If you plan to run MCP as a local process, you will use a command-line runtime such as UV (as shown in the examples) to launch the MCP service.

# Option A: Docker deployment (recommended) 
# 1) Create a project directory and pull the images
# 2) Start both TrendRadar and TrendRadar MCP services
# 3) Access the MCP endpoint at http://localhost:3333/mcp

# Start MCP service only (example)
docker compose up -d trendradar-mcp

# Start both TrendRadar and MCP services (example)
docker compose up -d trendradar trendradar-mcp

# Optional: run a single MCP container directly
docker run -d --name trendradar-mcp \
  -p 127.0.0.1:3333:3333 \
  -v $(pwd)/config:/app/config:ro \
  -v $(pwd)/output:/app/output:ro \
  -e TZ=Asia/Shanghai \
  wantcat/trendradar-mcp:latest

# Option B: Local runtime (STDIO/HTTP mix)
# If you prefer STDIO, ensure you have uv available and start the server as described in the MCP setup sections.

After starting the MCP service, you will have access to the following endpoints and capabilities depending on the deployment mode you chose. For HTTP mode, connect to http://localhost:3333/mcp. For STDIO mode, configure your client to launch the local MCP process and provide the proper command sequence as shown in the examples.

Configuration and usage notes

Key points to plan before you start pushing MCP analysis:

• Decide between HTTP (remote MCP) and STDIO (local MCP) deployment. HTTP is straightforward for remote hosts; STDIO is best when you want to host everything locally and use your editor or IDE integrations.
• The MCP server exposes tools for news querying, trending analysis, RSS queries, and various AI-assisted capabilities. You can enable or disable AI analysis, translation, or inspector tooling depending on your needs.

Security and access

Protect MCP access when deploying in production. If you expose the MCP HTTP endpoint, enable authentication or restrict network access to trusted hosts. For local STDIO setups, keep the MCP process within a secure environment and avoid exposing its internal endpoints publicly.

Troubleshooting and tips

If the MCP service cannot start, verify that the expected port (3333) is not in use and that required volumes (config and output) are mounted correctly. Check logs from the running container or process to identify startup errors and adjust environment variables (like TZ) as needed.

Notes on MCP clients and tools

This MCP supports a variety of clients that speak the MCP protocol, including editors and GUI tools. Tools exposed by the MCP service include querying latest news, searching by date, trending topic analysis, and deep-dive AI analysis like sentiment and cross-platform comparisons.

Supported MCP interaction methods

HTTP MCP: Access a remote MCP endpoint via a URL and send JSON-RPC-like requests over HTTP. STDIO MCP: Run the MCP server locally and connect clients via a local process, using a command like uv to launch the server inside the container or host environment.

Available tools

get_latest_news

Fetches the latest news entries from the configured data sources.

get_news_by_date

Retrieves news by a specific date range.

get_trending_topics

Provides trending topics across sources with optional filtering.

search_news

Search within the collected news for specific keywords.

find_related_news

Find related news across platforms for cross-referencing.

analyze_topic_trend

Analyze how a topic’s prominence changes over time.

analyze_data_insights

Generate data-driven insights from the news corpus.

analyze_sentiment

Assess sentiment across sources and platforms.

aggregate_news

Consolidate news records across platforms to remove duplicates.

compare_periods

Compare news metrics between different time periods.

generate_summary_report

Create a consolidated summary report of the analyzed data.

get_latest_rss

Fetch latest items from configured RSS feeds.

search_rss

Query RSS content for keywords.

get_rss_feeds_status

Check the status of configured RSS feeds.

get_current_config

Return current MCP server configuration.

get_system_status

Return basic system health information.

resolve_date_range

Parse and resolve natural language date ranges.