Integrations
Put built-environment forecasts inside the tools you already use.
Homecastr is the forecasting and data layer for applications, AI agents, notebooks, and research systems. Start with a live interface, then use the same public contract everywhere else.
One public contract
REST, MCP, schemas, and source clients stay aligned.
- OpenAPI operations
- 45
- MCP tools
- 44
- API version
- 1.12.0
- REST and MCP access
- 1 key
Available now
Live products and machine-readable interfaces
Each item below is available today. Registry claims are separated from source-only packages.
Live
Web app
Forecast map
Search an address, inspect forecast bands, and compare locations in the browser.
Open
OpenAPI 1.12.0
Application interface
REST API
45 documented operations for forecasts, data, benchmarks, status, and access.
Open
Streamable HTTP
AI agents
Remote MCP
44 structured tools for forecasts, data discovery, provenance, and evaluation.
Open
npm 1.16.0
Local agent clients
MCP npm package
Published stdio server for MCP-compatible editors, desktop assistants, and agent hosts.
Open
JSON
Code generation
OpenAPI contract
Machine-readable OpenAPI 3.1 contract for client generation and interface validation.
Open
Package status
Published means the registry serves it.
Source-ready packages are useful today, but they are not presented as registry releases.
Published
MCP server
Published on npm
Install with npx or connect directly to the hosted MCP endpoint.
Review package
Source ready
TypeScript client
Not published on npm
Typed client source and tests live in packages/sdk-typescript in the main repository.
Review package
Source ready
Python client
Not published on PyPI
The public client source is available on GitHub for source installation and review.
Review package
Built with Homecastr
First-party applications, demos, and research surfaces
These are live Homecastr surfaces, not placeholder partner projects or generated examples.
Application
Interactive forecast map
Address search, mapped forecast bands, saved places, and location comparison.
Explore
Application
Market pages
Browse public forecast summaries by market and move into the detailed map.
Explore
Demo
Forecast Open
Compare public market matchups, inspect the same board, and continue with API or MCP calls.
Explore
Research
Public data catalog
Inspect documented assets, access posture, coverage, provenance, and canonical schemas.
Explore
Research
Benchmark guide
Review public vintages, source manifests, and model-versus-baseline score tables.
Explore
Demo
Prompt library
Start from tested questions for places, comparisons, forecasts, and public data.
Explore
Start building
Small guides that lead into the complete reference
Call a forecast by addressREST quickstartBeginner5 minConnect an AI agentRemote MCP or npm stdio serverBeginner10 minChoose a forecast geographyAddress, parcel, tract, ZCTA, county, or stateBeginner10 minQuery standardized dataCatalog, coverage, schemas, and provenanceIntermediate15 minEvaluate against a benchmarkPublic vintages and score tablesIntermediate15 minUse a typed client from sourceTypeScript and Python package statusIntermediate10 min
Integration patterns
Choose the interface that matches the product.
One Homecastr API key can authenticate REST and MCP. The demo key remains available for public read evaluation.
Get a free API key| What you are building | Use | Why |
|---|---|---|
| Consumer or professional app | REST API | Address and geography forecasts with stable JSON contracts. |
| AI coding agent | MCP | Structured forecast, discovery, provenance, and benchmark tools. |
| TypeScript application | REST + source client | Use fetch directly or the tested in-repository typed client. |
| Python notebook | REST + source client | Use requests directly or install the public client from source. |
| Model evaluation | REST + artifacts | Query benchmark tables and retain version and provenance fields. |
| Interface validation | OpenAPI | Generate clients or compare implementation behavior to the public contract. |