# Sentinel Hub MCP for AI Agents AI Agent Connect > Sentinel Hub MCP lets you pull and process satellite imagery from missions like Sentinel and Landsat directly into your AI workflow. It handles STAC catalog searches, complex band combinations, and vegetation analysis without you having to download massive raw datasets first. Use it to turn raw Earth observation data into actionable maps and statistics in seconds. ## Overview - **Category:** cloud-infrastructure - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_xyVUwa3x77fWINQTr8feR7dKZks5UVMkLVCjWXd2/ai-agent-connect - **Tags:** satellite-imagery, earth-observation, geospatial-data, stac, remote-sensing, raster-processing ## Description You want to see what is happening on the ground without driving there or downloading gigabytes of raw data. This Connector connects your AI client to a professional satellite imagery pipeline. Instead of wrestling with raw files, you can ask your agent to find specific scenes based on cloud cover, location, or date. It handles the heavy lifting of processing the data in the cloud. For example, if you need to see how a forest is changing or where a new building went up, the AI can generate the right visual filters and run the math for you. You get back clean images or specific data points like vegetation density or moisture levels. It is a huge jump from manual GIS workflows because the AI understands the requests and handles the complex scripts behind the scenes. By using this through the Vinkius catalog, you get a direct line to high-quality Earth observation data. You can monitor crops, track urban growth, or assess wildfire damage in minutes rather than hours of manual processing. ## Tools ### catalog_search Search the STAC catalog for imagery using coordinates, date ranges, and specific collection IDs. This helps you find the right scenes quickly. ### check_sentinel_hub_status Verify your API connection and credentials are working correctly. Use this to troubleshoot your setup. ### generate_false_color_evalscript Create a script that highlights vegetation in red and urban areas in cyan. This is great for land cover maps. ### generate_ndvi_evalscript Get a script to color-code vegetation density from bare soil to dense forest. It makes plant health easy to see. ### generate_true_color_evalscript Create a script for natural-looking RGB satellite photos of any location. Use this for standard visual checks. ### get_catalog_collection Retrieve detailed information about a specific satellite data collection. This helps you understand what data is available. ### get_catalog_item Pull metadata for a specific item found in a catalog search. Use this to check details for a single scene. ### get_statistics Calculate mean, min, and max values over an area using a specific evalscript. This is perfect for time-series analysis. ### get_user_info Check your account profile and available usage quotas. Use this to manage your API limits. ### list_band_combinations See a list of predefined combinations like NDWI, SWIR, and Burn Severity. This saves you from writing complex scripts. ### list_catalog_collections Browse all available satellite data types like Landsat and Sentinel missions. This gives you a full view of the catalog. ### process_image Run an evalscript on a specific area to generate a processed satellite image. This is the core tool for creating custom maps. ### search_by_tile Find all scenes for a specific MGRS tile identifier within a date range. This is useful for standard spatial references. ### search_cloud_free Filter for satellite scenes that fall below your chosen cloud cover threshold. Use this to ensure you get clear data. ## Prompt Examples **Prompt:** ``` Look at the forest in the Amazon and tell me how much of it is actually green. ``` **Response:** ``` I'll generate an NDVI map for the Amazon basin. **Analysis Results:** - **Dense Vegetation:** 87% (Deep Green) - **Sparse Vegetation:** 8% (Yellow-Green) - **Water/Shadow:** 5% (Dark) This shows a high concentration of healthy forest cover in the targeted area. Would you like me to run a time-series analysis to see how this has changed over the last 12 months? ``` **Prompt:** ``` I need to see a clear image of Paris from last week with no clouds. ``` **Response:** ``` I'm searching for Sentinel-2 imagery over Paris (2.2, 48.8, 2.5, 48.9) from the last 7 days with less than 10% cloud cover. **Found Scenes:** - **Scene A:** 2.1% cloud cover (5 days ago) - **Scene B:** 7.3% cloud cover (3 days ago) I can generate a true-color image of Scene A for you now. Should I proceed? ``` **Prompt:** ``` What are the best ways to see where a fire burned recently? ``` **Response:** ``` To assess wildfire damage, I recommend using the **Burn Severity (NBR)** index. Here are the best band combinations for this: - **Burn Severity (NBR):** Combines NIR and SWIR bands to highlight burned vs. unburned areas. - **SWIR Composite:** Highlights burned areas in distinctive colors for broader context. - **False Color (Vegetation):** Good for distinguishing healthy plants from scorched earth. Would you like me to process one of these for a specific location? ``` ## Capabilities ### Search satellite scenes Find specific imagery based on your geographic coordinates, date range, and desired collection. ### Generate visual filters Create custom scripts to highlight specific features like vegetation, water, or urban areas. ### Calculate vegetation health Generate NDVI maps to see plant density and health across any specific region. ### Extract area statistics Get mean, min, and max values for vegetation or moisture indices over time. ### Filter for clear skies Automatically find the best available scenes that fall below your cloud cover threshold. ### Access curated indices Use predefined band combinations for things like burn severity and moisture levels. ## Use Cases ### Monitoring Crop Health An agricultural advisor asks for an NDVI map of a farm to check for drought stress using `generate_ndvi_evalscript`. ### Wildfire Damage Assessment An emergency manager uses `list_band_combinations` to find burn severity indices for a recent fire to assess damage. ### Urban Expansion Tracking A planner asks for mean vegetation stats over a city to see how park space is shrinking using `get_statistics`. ### Satellite Scene Discovery A researcher uses `catalog_search` to find the best images of a specific forest over three years for a study. ## Benefits - Get high-quality imagery without downloading raw data because this Connector processes everything in the cloud. - Generate instant vegetation maps with `generate_ndvi_evalscript` to see plant health clearly. - Filter out obstructed views using `search_cloud_free` to ensure you only see clear scenes. - Access standard indices like NDWI and SWIR quickly via `list_band_combinations`. - Track urban expansion and land changes over time with `get_statistics` for area analysis. - Create custom visual filters on the fly using `process_image` to highlight specific land features. ## How It Works The bottom line is you get instant access to processed satellite data without the overhead of managing raw files. 1. Subscribe to the Connector and register your account at the Copernicus Data Space. 2. Provide your OAuth2 client credentials in the required format. 3. Ask your AI client to search, process, or analyze satellite imagery for your project. ## Frequently Asked Questions **Can I use Sentinel Hub MCP to monitor crop health?** Yes, you can use it to generate NDVI maps that color-code vegetation density. This helps you see plant health across different fields and regions instantly. **How does Sentinel Hub MCP help with wildfire damage?** It provides access to burn severity indices and SWIR composites. Your agent can use these to identify exactly which areas were impacted by a fire. **Can I find cloud-free images of a specific city?** Yes, the Connector can filter the satellite catalog for scenes that fall below a specific cloud cover percentage, ensuring you get clear images for your analysis. **Does Sentinel Hub MCP support Landsat data?** It supports multiple missions, including Sentinel and Landsat. You can switch between these data collections depending on your specific needs. **Can I get statistics on vegetation density?** Yes, you can calculate mean, min, max, and standard deviation for vegetation indices over any area of interest with temporal aggregation. **Is Sentinel Hub MCP good for urban planning?** It is excellent for urban planning because it allows you to analyze land use changes and monitor urban expansion through multi-temporal statistical analysis. **What is an evalscript and how do I use one?** An evalscript is a small JavaScript program that tells Sentinel Hub how to process satellite bands into an output image. It defines which bands to use, how to combine them, and what colors to assign. You can use the generate_ndvi_evalscript or generate_true_color_evalscript tools to get ready-made evalscripts, then pass them to the process_image tool. **Can I analyze vegetation health with this server?** Absolutely. Generate an NDVI evalscript with the generate_ndvi_evalscript tool, then process imagery for your area of interest with the process_image tool. For time-series analysis, use the get_statistics tool with temporal aggregation to track vegetation changes over weeks or months. The search_cloud_free tool helps you find clean scenes without cloud contamination. **What is the difference between this server and the Copernicus Data Space server?** The Copernicus Data Space server focuses on product catalogue search and download — finding and retrieving raw satellite data files. Sentinel Hub focuses on on-the-fly processing — rendering images, computing indices, and generating statistics without downloading raw data. They complement each other: use Copernicus for data discovery and bulk download, Sentinel Hub for real-time analysis and visualization.