# Altos [Health: Active]

**Category:** 🏠 Real Estate  
**Repository:** https://github.com/pipeworx-io/mcp-altos  
**GitHub Stars:** 0  
**npm Downloads (last month):** 408  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/altos

## Description
Altos Research MCP — Real estate market intelligence

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "altos": {
    "command": "npx",
    "args": ["-y","@pipeworx/mcp-altos"]
  }
}
```

## Documentation

## What Altos does

The Altos MCP server provides access to weekly U.S. real-estate market data from Altos Research. It focuses on active listing conditions rather than completed-sale prices, including median list price, price reductions, days on market, inventory, and weekly movement. Results can be requested for the national market, a state, a metropolitan area, or a ZIP code.

The data is useful for questions such as whether a market is becoming more competitive, how quickly listings are moving, and how inventory differs between regions. It belongs in the Real Estate category and is particularly relevant to housing reports, market briefs, lender analysis, and comparisons between peer metros.

## How it works

A request supplies an Altos region identifier to the `altos_market_stats` tool. Region codes use formats such as `us_national`, `us-co`, `us-co-denver`, and `us-co-denver-80202`. Optional filters select `single_family` or `condo` properties and one of the four market quartiles. The first quartile represents the lower-priced segment, while the fourth represents the upper-priced segment.

Altos generally publishes data weekly, usually early in the week for the preceding week. Pipeworx applies a 24-hour cache, so repeated requests during the same period may return the same values. The latest reporting period can be revised as additional listings arrive.

The Altos MCP server can run as a scoped Pipeworx endpoint or as a local package communicating over standard input and output. The gateway version also exposes shared Pipeworx tools, including natural-language routing and tool discovery. The local version exposes only the Altos pack tools.

## Setup and configuration

For a remote MCP connection, configure the Pipeworx endpoint at `https://gateway.pipeworx.io/altos/mcp`. You can instead connect to `https://gateway.pipeworx.io/mcp` to access the full gateway directly. Both routes use the Pipeworx service; the scoped endpoint lists the Altos tools directly and also includes shared gateway tools.

For local use, configure an MCP client to run `npx -y @pipeworx/mcp-altos`. This option does not require a Pipeworx account or a gateway round trip. The package communicates over stdin and stdout and responds to MCP initialization, tool-listing, and tool-call requests.

Authentication has two supported paths. Supply an Altos partner key in the `_altosKey` argument when using your own commercial Altos relationship, or use the Pipeworx Housing Vertical subscription without that argument. HTTP callers can invoke the gateway tool endpoint directly with JSON.

## Tools and capabilities

The primary documented pack tool is `altos_market_stats`. It supports:

- Weekly market statistics for national, state, metro, and ZIP regions
- Filtering by single-family or condominium property type
- Filtering by first through fourth market quartile
- Regional comparisons by making multiple requests
- Access through MCP clients or the Pipeworx HTTP API

Gateway connections additionally provide shared tools such as `ask_pipeworx`, `discover_tools`, `search_within`, `remember`, and `recall`. These shared tools are not included by the standalone local package.

## Limitations and notes

Altos primarily measures active listings. Its median price should not be interpreted as a completed-sale price, and sale-price changes generally come from other sources such as county-record data. Active inventory is also not the same as months of supply; calculating that measure requires a monthly sales or absorption rate.

Condominium coverage varies by market and may be unavailable in rural areas. ZIP-level results can be volatile because small listing counts produce noisier weekly changes. Quartile boundaries are recalculated weekly, so the same quartile does not necessarily represent a fixed price range over time. Treat the newest weekly value as provisional rather than using it alone for production decisions.

The Altos MCP server is an independent, unofficial integration and is not affiliated with or endorsed by Altos Research. The package is released under the MIT license.

_Full upstream README: https://allmcps.com/mcp/altos/readme_