# World Monitor [Health: Active]

**Category:** 📊 Monitoring  
**Repository:** https://github.com/koala73/worldmonitor  
**GitHub Stars:** 87154  
**npm Downloads (last month):** 404  
**Views:** 5  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/world-monitor

## Description
Live global intelligence: real-time markets, conflicts, country risk, chokepoints, energy. 39 tools.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "world-monitor": {
    "command": "npx",
    "args": ["-y","worldmonitor"]
  }
}
```

## Documentation

## What World Monitor MCP server does

The World Monitor MCP server gives agents programmatic access to the information behind the World Monitor global intelligence dashboard. Its coverage includes curated global and regional news, geopolitical developments, conflicts, country risk, infrastructure, chokepoints, energy, financial markets, commodities, crypto, and disaster-related signals.

The project also describes a Country Instability Index with live scores, risk bands, and approximate 24-hour movement for 31 Tier-1 countries. Finance-oriented coverage includes stock exchanges, commodities, crypto, and a market composite. The current listing identifies 39 tools, making the server suitable for agents that need to retrieve several kinds of global signals through one MCP interface.

## How it works

The World Monitor MCP server is hosted at `https://worldmonitor.app/mcp` and uses Streamable HTTP. Clients can request the available tool definitions without authentication through `tools/list`. Executing a tool through `tools/call` requires either an `X-WorldMonitor-Key` header or OAuth authentication.

The underlying World Monitor project combines news aggregation, AI-generated briefs, map-based monitoring, and cross-stream correlation across military, economic, disaster, and escalation signals. The MCP endpoint provides an agent-facing access path to this broader system rather than requiring an agent to operate the dashboard manually. The server also publishes Agent Skills through the draft `io.modelcontextprotocol/skills` extension.

## Setup and configuration

Configure an MCP client with the hosted endpoint:

```text
https://worldmonitor.app/mcp
```

Authentication must be configured for tool execution using the supported World Monitor key header or OAuth flow. The supplied material does not specify how to obtain a key, define OAuth credentials, or configure individual client products, so those details should be confirmed in the project’s current documentation.

The repository can also be run as a local World Monitor application with Node.js. Its documented development flow is to clone the repository, run `npm install`, start it with `npm run dev`, and open the local development address. That quick-start procedure is for the dashboard project; it is not presented as a separate local MCP launch command.

## Tools and capabilities

The available capability areas include:

- Retrieve curated news across global and regional categories.
- Work with AI-synthesized news briefs.
- Inspect country instability scores and recent movement.
- Access market, commodity, crypto, and energy-related intelligence.
- Examine conflict, military, disaster, infrastructure, and chokepoint signals.
- Correlate signals across economic, military, disaster, and escalation streams.

These capabilities fit research, risk monitoring, situation reports, market context, and agent workflows that need current events combined with structured global indicators.

## Limitations and notes

The endpoint requires authentication for tool calls even though tool discovery is public. The material does not provide a complete per-tool catalog, schemas, rate limits, retention guarantees, or freshness guarantees for every data source. Results may therefore vary by source and feature.

World Monitor also offers web, desktop, and self-hosted application surfaces, but those should not be assumed to have identical configuration or access behavior to the MCP endpoint. Feature-specific data sources in the broader application may require credentials. Local AI support through Ollama is described for the application and does not establish that every MCP capability works without service authentication.

_Full upstream README: https://allmcps.com/mcp/world-monitor/readme_