Public Hosted Server: https://gdelt.caseyjhand.com/mcp

Tools

Nine tools across two GDELT APIs — DOC API for global print/web news (last 3 months, 65 languages, no auth) and TV API for US television transcripts (2009–Oct 2024, 150+ stations):

gdelt_search_articles

Search the last 3 months of global news with GDELT's full query syntax.

• Keywords, phrases ("bird flu"), boolean OR, and exclusion (-sports)

• Filter operators: sourcecountry:, sourcelang:, domain:, theme: (GKG taxonomy), tone</tone>

• Proximity and repetition: near20:"flu virus", repeat3:"outbreak"

• Configurable sort (relevance, dateDesc, dateAsc, toneDesc, toneAsc, hybridRel) and result count (up to 250)

• Returns URL, title, publication date, domain, language, source country, and social image URL

• 250 is a hard per-call ceiling, not a page size — GDELT exposes no cursor. Fill it and the response returns continuationWindows: the queried window halved, ready to re-query. The halves overlap by a second so nothing falls through the seam; de-duplicate by url

• Query is echoed in response for chaining

gdelt_get_coverage_timeline

Retrieve when coverage of a topic spiked, with three modes:

• volume — normalized percentage of all global coverage per timestep

• volume_with_articles — volume plus top articles driving each spike; use for signal detection without a follow-up search call

• tone — average sentiment score per timestep (combine with gdelt_get_tone_distribution for the full picture)

• Every article reference is always in structuredContent; the text surface renders the first 3 links per timestep beside that timestep's true count, and points: ["<date>"] renders named timesteps in full

• Configurable smoothing and time range

gdelt_get_tone_distribution

Snapshot tone histogram across all articles matching a query.

• Bins from approximately −30 to +30; each bin includes representative article URLs

• Summary fields: peakNegativeBin, peakPositiveBin, neutralPct (% of articles in the −2 to +2 range)

• Distinct from the tone timeline — distribution across all matching articles, not over time

gdelt_get_coverage_breakdown

Multi-series time series showing which countries or languages drove coverage.

• Break down by language or country

• Top 10 series by total volume; remaining series aggregated into an "Other" bucket and named in otherSeriesLabels

• Nothing is dissolved into "Other" anonymously — pass any label to series: ["<label>"] and that series comes back complete under selectedSeries, alongside the usual overview

• Values are normalized — the topic's share of media output, not absolute article counts. Small media markets with concentrated coverage rank above large markets with diverse output, so a high value means the topic dominated that source's coverage rather than that it published the most articles

• Use to trace how a story propagated geographically

gdelt_search_tv

Search US television news transcripts (2009–Oct 2024) with per-station airtime analysis.

• Up to 10 structured stations (e.g. ["CNN", "FOXNEWS"]) — or place a station: selector directly in the query

• Normalize query-matching coverage to relative airtime % or return raw matching 15-second clip counts

• Optional dateres aggregation: hour, day, week, month, or year; returned timestamps are normalized to ISO 8601

• Responses select at most 500 points per page in deterministic date-then-station order, then group them by station. Use nextOffset with the same query inputs to retrieve the next page; content[] renders every point in the structured page

• TV-specific operators: market:, show:, context:

• Use gdelt_list_tv_stations to verify station active date ranges before querying recent events

gdelt_get_tv_clips

Retrieve actual TV news clips driving a coverage signal.

• Up to 3,000 clips per call

• Each clip: show name, station, air timestamp, 15-second transcript excerpt, direct Archive.org link, and optional thumbnail

• 3,000 is a hard per-call ceiling, not a page size — GDELT exposes no cursor. Fill it and the response returns continuationWindows: the queried window halved, ready to re-query. The halves overlap by a second so nothing falls through the seam; de-duplicate by archiveUrl

• Sort by relevance, date descending, or date ascending

gdelt_get_tv_context

Vocabulary framing analysis for TV coverage of a topic.

• Returns the most frequent non-stopword terms from matching clips

• Relative frequency scores (query term = 100)

• Use to identify narrative framing, related concepts, or follow-up search terms

gdelt_get_tv_trending

Zero-argument entry point for the current TV news cycle.

• Returns trending topics, keywords, and phrases dominating national networks

• Updated every 15 minutes

• Note: coverage data ends Oct 2024; results reflect the archive endpoint, not a live feed

gdelt_list_tv_stations

Station metadata lookup before querying.

• All available stations with market, network, monitoring start date, and end date

• isActive flag — true when end date is within the last 24 hours

• Use to verify a station was active during a target time period

Features

Built on @cyanheads/mcp-ts-core:

• Declarative tool definitions — single file per tool, framework handles registration and validation

• Unified error handling — handlers throw, framework catches, classifies, and formats

• Pluggable auth: none, jwt, oauth

• Swappable storage backends: in-memory, filesystem, Supabase, Cloudflare KV/R2/D1

• Structured logging with optional OpenTelemetry tracing

• STDIO and Streamable HTTP transports

GDELT-specific:

• Shared rate-limit queue (1 req/5s) across all tools — enforces GDELT's published limit without caller coordination

• Two service layers (GdeltDocService, GdeltTvService) mapping clean tool parameters to the DOC and TV API URL conventions

• TV station filter operators embedded in query strings internally — callers pass structured stations arrays, not raw query syntax

Agent-friendly output:

• Query echo on every response — searches return the original query and applied timespan so agents can chain calls without re-deriving parameters

• Discriminated series labels — timeline and breakdown responses carry typed label fields ("Volume Intensity", "Average Tone", language/country names) rather than positional arrays

• Structured station metadata — isActive boolean and ISO 8601 date fields let agents reason about TV station availability without parsing date strings

• Partial-coverage signals in distribution output — neutralPct, peakNegativeBin, peakPositiveBin summary fields let agents branch on sentiment without histogramming the raw bins themselves

Getting started

Public Hosted Instance

A public instance is available at https://gdelt.caseyjhand.com/mcp — no installation required. Point any MCP client at it via Streamable HTTP:

{
  "mcpServers": {
    "gdelt-mcp-server": {
      "type": "streamable-http",
      "url": "https://gdelt.caseyjhand.com/mcp"
    }
  }
}

Self-Hosted / Local

Add the following to your MCP client configuration file.

{
  "mcpServers": {
    "gdelt-mcp-server": {
      "type": "stdio",
      "command": "bunx",
      "args": ["@cyanheads/gdelt-mcp-server@latest"],
      "env": {
        "MCP_TRANSPORT_TYPE": "stdio",
        "MCP_LOG_LEVEL": "info"
      }
    }
  }
}

Or with npx (no Bun required):

{
  "mcpServers": {
    "gdelt-mcp-server": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@cyanheads/gdelt-mcp-server@latest"],
      "env": {
        "MCP_TRANSPORT_TYPE": "stdio",
        "MCP_LOG_LEVEL": "info"
      }
    }
  }
}

Or with Docker:

{
  "mcpServers": {
    "gdelt-mcp-server": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-e", "MCP_TRANSPORT_TYPE=stdio",
        "ghcr.io/cyanheads/gdelt-mcp-server:latest"
      ]
    }
  }
}

For Streamable HTTP, set the transport and start the server:

MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 bun run start:http
# Server listens at http://localhost:3010/mcp

Prerequisites

• Bun v1.3.0 or higher (or Node.js v24+).

• No API key required — GDELT is a free public API.

Installation

1. Clone the repository:

git clone https://github.com/cyanheads/gdelt-mcp-server.git

2. Navigate into the directory:

cd gdelt-mcp-server

3. Install dependencies:

bun install

4. Configure environment:

cp .env.example .env
# edit .env if you need to override defaults

Configuration

All configuration is validated at startup via Zod schemas in src/config/server-config.ts. Key environment variables:

See .env.example for the full list of optional overrides.

Running the server

Local development

• Build and run:

  # One-time build
  bun run rebuild
  
  # Run the built server
  bun run start:stdio
  # or
  bun run start:http

• Run checks and tests:

  bun run devcheck   # Lint, format, typecheck, security audit
  bun run test       # Vitest test suite
  bun run lint:mcp   # Validate MCP definitions against spec

Docker

docker build -t gdelt-mcp-server .
docker run --rm -p 3010:3010 gdelt-mcp-server

The Dockerfile defaults to HTTP transport, stateless session mode, and logs to /var/log/gdelt-mcp-server. OpenTelemetry peer dependencies are installed by default — build with --build-arg OTEL_ENABLED=false to omit them.

Project structure

Development guide

See CLAUDE.md for development guidelines and architectural rules. The short version:

• Handlers throw, framework catches — no try/catch in tool logic

• Use ctx.log for request-scoped logging, ctx.state for tenant-scoped storage

• Register new tools via the barrels in src/mcp-server/tools/definitions/index.ts

• Wrap GDELT API calls: validate raw JSON → normalize to domain type → return output schema; never fabricate missing fields

Contributing

Issues and pull requests are welcome. Run checks and tests before submitting:

bun run devcheck
bun run test

License

Apache-2.0 — see LICENSE for details.