## Quick Start

1. Add the Slingshot MCP server to your AI client configuration.
2. Configure authentication with your API key.
3. Start querying space data through natural language.

## Server Configuration

Add the Slingshot MCP server to your MCP settings. The server supports SSE (Server-Sent Events) and stdio transports. The JSON below is the same shape other MCP clients expect; this guide uses Cursor as the primary example.

```json
{
  "mcpServers": {
    "slingshot": {
      "url": "https://api.portal.slingshot.space/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}
```

#### Cursor

Add this to your MCP settings file:

macOS:`~/Library/Application Support/Cursor/User/globalStorage/mcp.json`  
Windows:`%APPDATA%\Cursor\User\globalStorage\mcp.json`

## Authentication

The MCP server uses the same Bearer token authentication as the REST API. Include your API key in the `Authorization` header.

### Recommended: Create a Dedicated API Key

We strongly recommend creating a separate API key specifically for AI agents. Natural language queries can result in unpredictable API call patterns, and a dedicated key allows you to monitor usage, set appropriate rate limits, and track costs independently from your other integrations.

### Security Note

Never commit API keys to version control. Use environment variables or secure secret management for production deployments.

## Available Tools

The MCP server exposes 53 tools organized by API group. All tools follow the same authentication and error handling patterns as the REST API.

### Space Object Catalog (SGSN)

Catalog endpoints for space object summaries and orbital products: TLE, observation tracks and their export products (TDM, image chips), and predicted ephemeris.

7 tools

- `list_space_object_summaries`
- `list_tracks`
- `get_track_tdm_export`
- `get_track_image_chips_export`
- `get_latest_tle`
- `get_tle_by_date`
- `get_predicted_ephemeris`

### Space Contextual Data (Seradata)

Space Contextual Data API: deep technical and operational context for spacecraft, rocket bodies, debris, organizations, launch vehicles, and subsystems. Powered by Seradata.

44 tools

- `get_spacecraft`
- `list_spacecraft`
- `list_spacecraft_aliases`
- `list_rocket_bodies`
- `get_rocket_body`
- `list_debris`
- `get_debris`
- `list_countries`
- `get_country`
- `list_companies`
- `get_company`
- `list_launch_sites`
- `get_launch_site`
- `list_launch_vehicles`
- `list_launch_vehicle_stages`
- `get_launch_vehicle_stages`
- `get_launch_vehicle_spec`
- `list_launches`
- `get_launch`
- `list_engines`
- `get_engine`
- `list_solar_arrays`
- `get_solar_array`
- `list_batteries`
- `get_battery`
- `list_antennas`
- `get_antenna`
- `list_thrusters`
- `list_events`
- `get_event`
- `list_communication_payloads`
- `get_communication_payload`
- `list_optical_payloads`
- `get_optical_payload`
- `list_radar_payloads`
- `get_radar_payload`
- `list_navigation_payloads`
- `get_navigation_payload`
- `list_early_warning_payloads`
- `get_early_warning_payload`
- `list_sigint_payloads`
- `get_sigint_payload`
- `list_scientific_or_other_payloads`
- `get_scientific_or_other_payload`

### Marketplace

Marketplace API for tracking orders and listing the data products. All transactions are metered in SPU (Slingshot Processing Units). Available now: retrieve your orders (get my orders) and place orders (create order). Coming soon: product catalog—list products and get product details to browse offerings before ordering.

1 tools

- `get_my_orders`

### Space Object Analytics

The Space Object Analytics API detects and analyzes space object behavior: out-of-family (cohort) anomalies; maneuver, drift, photometric, orbital element, and close-approach reports; and per-object history for drill-down. Out-of-family reports are published once per UTC day per orbit regime (LEO or GEO are separate snapshots); a single-band fleet typically needs one daily GET with the matching regime—see that operation for cadence and caching.

1 tools

- `get_out_of_family_report`

## Example Usage

Once configured, you can ask your AI assistant natural language questions that will be automatically translated into MCP tool calls.

### Natural Language Query

"Show me the latest TLE for the ISS"

MCP TOOL CALL 
`list_space_object_summaries(norad_ids="25544" or comma-separated NORADs) -> get_latest_tle(space_object_id from catalog response)`

### Natural Language Query

"What payloads does the ISS (NORAD ID 25544) have?"

MCP TOOL CALL 
`list_spacecraft(norad_id="25544") -> get_spacecraft(space_object_id from list) -> list_optical_payloads(seradata_spacecraft_id) + list_communication_payloads(seradata_spacecraft_id)`

### Natural Language Query

"What objects has Slingshot flagged for anomalies in LEO today?"

MCP TOOL CALL 
`get_out_of_family_report(orbit_regime="LEO") (one daily snapshot per regime; optional comma-separated norad_ids)`

## Additional Resources

[**Model Context Protocol Specification** Official MCP documentation and protocol details](https://modelcontextprotocol.io/specification)  
[**MCP overview** Introduction to Model Context Protocol, transports, and connecting clients to remote MCP servers](https://modelcontextprotocol.io/)