Get Started in Three Steps
Follow this guide to create your account, explore global events, and make your first API call.Step 1: Create Your Account
Sign up
Visit gdeltcloud.com/auth/sign-up and create your account with email and password.
Access the dashboard
Navigate to gdeltcloud.com/dashboard to access your account.
Step 2: Try the Research Agent
Open the Research Agent
From your dashboard, click on Research in the sidebar or navigate to the Research page.
Ask your first question
Try asking a question in natural language, such as:
The Research Agent will find relevant media event stories and present them with article links, entity context, and sentiment analysis.
Review the results
Results include media event clusters (stories), the articles that make up each story, linked entities, and CAMEO event metrics like Goldstein scale and tone scores.
Data starts from January 2025 and is growing over time as historical data is gradually added. New data is updated hourly.
Step 3: Make Your First API Call
For programmatic access, you’ll need an API key on the Analyst or Professional plan.Generate an API key
Go to Settings → API Keys and click Create New Key.
- Give your key a descriptive name (e.g. “Development Key”)
- Click Generate
- Copy the key immediately — it is shown only once
Next Steps
Explore the API
Learn about all available API endpoints, filters, and response detail levels.
Connect AI Agents
Use the MCP server to connect Claude, Cursor, or custom agents to GDELT data.
Research Agent
Learn how the AI-powered Research Agent helps you explore global events.
Common Questions
What data is available?
What data is available?
GDELT Cloud provides access to GDELT event data starting from January 2025, with new data updated hourly. Historical coverage is being expanded over time. The data includes:
- Media Events: News articles clustered into stories with sentiment, actors, and locations
- Entities: People and organizations linked across stories via Wikipedia (GEG entities)
- Domains: News source profiles with coverage patterns and top entities
How is the data organized?
How is the data organized?
GDELT Cloud organizes data in layers:
- Articles: Individual news articles from thousands of global sources
- Events: Structured CAMEO-coded events extracted from articles (who did what to whom, where)
- Media Events: Clusters of related articles forming a single “story” — the primary unit of analysis
- Entities: People and organizations identified across stories, linked to Wikipedia for disambiguation
What are the detail levels?
What are the detail levels?
API responses support progressive disclosure via the
detail parameter:summary: Compact headline cards (~120 tokens/cluster) — great for scanningstandard: Adds article links and entity details (~500 tokens/cluster)full: Complete data with per-article metrics (~1,000 tokens/cluster)
summary for discovery, then zoom into specific stories.Can I export data?
Can I export data?
Yes! All query results are returned as JSON via API responses. The Research Agent also displays results in structured formats with source links.
Need help? Check our API Reference for detailed documentation.

