MCP Carvana
# @striderlabs/mcp-carvana
MCP (Model Context Protocol) connector for [Carvana](https://www.carvana.com) — the online car marketplace. Enables AI assistants to help users search for vehicles, estimate financing, check trade-in values, and more.
## Installation
```bash
npm install @striderlabs/mcp-carvana
```
Or run directly with npx:
```bash
npx @striderlabs/mcp-carvana
```
## Configuration
Add to your MCP client configuration (e.g. Claude Desktop `claude_desktop_config.json`):
```json
{
"mcpServers": {
"carvana": {
"command": "npx",
"args": ["-y", "@striderlabs/mcp-carvana"]
}
}
}
```
## Available Tools
### `search_vehicles`
Search Carvana inventory with filters for make, model, year, price, mileage, body style, drivetrain, fuel type, and more.
### `get_vehicle_details`
Get full details for a specific vehicle by stock number or VIN.
### `get_financing_estimate`
Calculate estimated monthly payments based on vehicle price, down payment, credit tier, and loan term.
### `check_delivery_availability`
Check whether Carvana delivers to a specific zip code and get estimated timeframes.
### `get_vehicle_history`
Retrieve vehicle history report summary including accidents, owners, and service records.
### `get_trade_in_value`
Get an estimated trade-in offer for your current vehicle based on make, model, year, mileage, and condition.
### `find_nearby_vending_machines`
Find Carvana car vending machine pickup locations near a city or zip code.
### `compare_vehicles`
Compare 2–4 vehicles side-by-side with a structured breakdown of key dimensions.
### `search_by_budget`
Find vehicles that fit within a target monthly payment budget, with financing assumptions.
### `get_market_insights`
Get market context for a specific make/model including pricing trends and inventory levels.
## Example Usage
**Search for a used Toyota RAV4 under $30,000:**
> "Find me a used Toyota RAV4 under $30,000 on Carvana"
**Estimate monthly payment:**
> "What would my monthly payment be for a $25,000 car with $3,000 down over 60 months with good credit?"
**Trade-in value:**
> "What's my 2019 Honda Accord with 45,000 miles worth as a trade-in?"
**Budget search:**
> "What cars can I get on Carvana for $400/month with $2,000 down?"
## Development
```bash
# Install dependencies
npm install
# Build
npm run build
# Pack for distribution
npm pack
```
## License
MIT
TDQS
Scored across 10 tools
Every tool has a clearly distinct purpose with no overlap. For example, 'check_delivery_availability' focuses on delivery logistics, 'compare_vehicles' on side-by-side comparisons, and 'search_vehicles' on general filtering, making misselection unlikely.
All tool names follow a consistent verb_noun pattern with clear, descriptive verbs like 'check', 'compare', 'find', 'get', and 'search'. This uniformity enhances predictability and readability.
With 10 tools, the server is well-scoped for a car-buying domain, covering key workflows from search and comparison to financing and logistics. Each tool earns its place without feeling excessive or insufficient.
The toolset provides complete coverage for the car-buying process, including search, details, comparison, financing, trade-ins, market insights, delivery, and pickup locations. No obvious gaps exist that would hinder agent workflows.