# Earth Imagery MCP for NASA EPIC data. AI Agent Connect

> Earth Imagery MCP gives your AI client direct access to the NASA EPIC camera aboard the DSCOVR spacecraft. Located a million miles away at the L1 Lagrange point, this camera captures the entire sunlit side of Earth at once. You can pull the latest natural color frames, browse historical archives by date, or find specific scans centered on a precise latitude and longitude without needing any API keys.

## Overview
- **Category:** weather-climate
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_m7lRH6YcszSdTIc8pyo6WnQQS4zP0V5pDS35lHf1/ai-agent-connect
- **Tags:** nasa, epic, dscovr, earth-imagery, satellite-imagery, earth-observation, remote-sensing, space, true-color, whole-earth

## Description

You can now give your AI agent a view of the entire sunlit Earth. This MCP connects your client to the NASA EPIC camera on the DSCOVR spacecraft, which sits at the Earth-Sun L1 point. Because it stays positioned a million miles away, it sees the whole sunlit disc in a single frame. You can use it to monitor daylight cloud cover, track dust plumes, or observe volcanic ash movements. The data comes in two flavors: natural color for true-to-life calibration and enhanced color for better contrast and cloud detail. Whether you need the absolute freshest frame published today or a specific scan from years ago, your agent can fetch the exact image, its capture time, and the precise coordinates of the spacecraft, Moon, and Sun.

## Tools

### get_earth_images_by_date
Retrieves all Earth images captured by NASA EPIC on a specific ISO date. It provides the capture time, sub-spacecraft point, and a permanent archive URL for every scan from that day.

### get_latest_earth_images
Fetches the most recent natural color images published by the EPIC camera. This tool returns the newest frames first, including their capture time and stable archive URLs.

### list_epic_collections
Lists the available NASA EPIC imagery collections, including natural and enhanced color sets. It shows how many dates each collection covers and identifies the most recent available date.

### list_epic_dates
Provides a paginated list of all dates for which NASA EPIC has captured imagery. Use this to find valid dates before requesting specific images.

### find_earth_images_near
Locates EPIC scans that were centered closest to a specific latitude and longitude. It returns the distance and bearing from the target location to the scan center.

### get_earth_image
Retrieves a single NASA EPIC frame using a specific identifier. The response includes the archive URL, capture time, sub-spacecraft point, and the J2000 positions of the spacecraft, Moon, and Sun.

## Prompt Examples

**Prompt:** 
```
Show me the most recent photo of Earth.
```

**Response:** 
```
EPIC's natural collection, captured 2026-09-17 00:03:42 UTC with scan centre 6.5N, 178.4E: https://epic.gsfc.nasa.gov/archive/natural/2026/09/17/png/epic_1b_20260917000831.png
```

**Prompt:** 
```
Was New York facing the EPIC camera recently?
```

**Response:** 
```
The closest recent natural scan to 40.71N, 74.01W is epic_1b_20260916163014, 3859 km away on a bearing of 168 degrees. The scan centre was south of the city, so New York sat in the upper part of that frame: https://epic.gsfc.nasa.gov/archive/natural/2026/09/16/png/epic_1b_20260916163014.png
```

**Prompt:** 
```
Get all images from July 6, 2015.
```

**Response:** 
```
On 2015-07-06, EPIC took five frames spaced hours apart. The first, epic_1b_20150706051132 at 05:01 UTC, was centered on 19.1N 112.7E. You can access the full set through the NASA archive.
```

## Capabilities

### Real-time monitoring
Your agent uses the latest natural color frames to check current daylight conditions.

### Geospatial lookups
The AI finds specific imagery by matching a latitude and longitude to the nearest scan center.

### Historical research
Your agent browses the full archive of capture dates to find specific past events.

### Atmospheric analysis
The AI accesses enhanced color collections to inspect cloud detail and dust plumes.

### Orbital positioning
Your agent retrieves the exact J2000 positions of the spacecraft, Moon, and Sun for any frame.

## Use Cases

### Cloud Cover Tracking
An agent monitors the natural color collection to observe how cloud patterns move across the sunlit Earth.

### Volcanic Ash Monitoring
Researchers use the enhanced color frames to get better contrast for detecting ash plumes in the atmosphere.

### Geographic Event Verification
A user asks if a specific city was visible in the most recent satellite pass to verify local conditions.

### Historical Data Retrieval
An AI agent pulls a sequence of frames from a specific date in the past to study Earth's rotation.

## Benefits

- Access NASA's full-disc imagery without managing API keys.
- Get true-color or enhanced-contrast images depending on your analysis needs.
- Locate specific views using simple latitude and longitude coordinates.
- Pull data from a spacecraft positioned a million miles away via a single connection.

## How It Works

Connecting to this data is a one-time setup that gives your AI client immediate access to NASA's archive.

1. Connect your AI client to the Earth Imagery MCP via Vinkius.
2. Ask your agent to find the latest images or a specific date.
3. The MCP fetches the requested frames from the DSCOVR spacecraft data.
4. Your agent receives the image URLs, timestamps, and coordinates.
5. The images are displayed or processed directly in your client.

## Frequently Asked Questions

**Do I need a NASA API key to use this?**
No. This MCP is keyless and handles all connections to the NASA EPIC data automatically.

**What is the difference between natural and enhanced collections?**
Natural color is the calibrated true-color version released the same day. Enhanced color has boosted contrast for better cloud detail and lags behind the natural collection by days or weeks.

**How often does the camera take pictures?**
The camera captures roughly one frame per hour across the sunlit side of the Earth.

**Can I find images of a specific location?**
Yes. You can provide a latitude and longitude, and the AI will find the frames where that location was closest to the center of the view.

**Which AI clients can use this MCP?**
Any MCP-compatible client like Claude, Cursor, Windsurf, or VS Code can use this tool once connected through Vinkius.

**Do I need an API key?**
No. This reads NASA's public EPIC archive, which requires no key or account.

**What is the difference between the natural and enhanced collections?**
They are the same frames. Natural is the calibrated true-colour composite and publishes the same day; enhanced has colour and contrast boosted to show cloud and surface detail, and its pipeline trails natural by days to weeks. Ask for natural when freshness matters.

**Why is there no imagery on the date I asked for?**
DSCOVR does not image every day. The tool reports the dates that do have frames — call list_epic_dates first when you are unsure, and switch collections if one has nothing for that day.

**What does the distance in find_earth_images_near mean?**
Great-circle kilometres from the place you gave to the scan centre — the point on Earth directly under the spacecraft. A small distance means the place was near the middle of the frame and facing the camera; EPIC sees each longitude roughly once per day.