Meet Geo™ — Your AI Property Analyst
Data version: Q2 2026 · Last updated 2026-06-23
TL;DR. Geo™ is the AI analyst layer built into every Buildability™ Report. Ask any question about any U.S. property in plain English — "Can I build a duplex on this lot?", "What is the flood risk for this address?", "How much can I build here under the current zoning?" — and Geo™ returns a cited, actionable answer in seconds. Geo™ is powered by Anthropic Claude — routed across Haiku 4.5, Sonnet 4.6, and Opus 4.7 by question complexity — and pulls from all 20+ Buildability™ data sources.
How Geo™ works
Geo™ is an AI agent that has live access to Buildability™'s full data layer: zoning databases, FEMA flood maps, EPA contamination records, county GIS, comparable sales, permit histories, and more. When you ask a question, Geo™ identifies the relevant data sources, pulls the raw values, reasons over them with Anthropic Claude, and returns a cited answer with direct references to the source records. You can follow up with deeper questions — Geo™ maintains context across a conversation about a specific parcel.
What you can ask
Geo™ handles the full range of property questions: "What's my ADU size limit?", "Is this a flood zone?", "What's the maximum building height?", "Are there any environmental red flags within a mile?", "What comparable parcels sold for nearby?", "What's the permit timeline estimate for this project?", "Can I convert the garage?", "What's the parking requirement?", "Are there any pending zoning changes in this area?", "What's the setback on the side yard?". Geo™ answers every question with citations to the underlying government source.
One model family, routed by complexity
Geo™ runs end-to-end on Anthropic Claude: Opus 4.7 (1M-token context with adaptive thinking) for the highest-stakes complex reasoning, Sonnet 4.6 for the broader complex tier, and Haiku 4.5 for fast lookups — with a complexity router picking the cheapest tier that can answer correctly. Every answer is grounded in the cited government source for that parcel; when the source data is incomplete or ambiguous, Geo™ flags the finding as low-confidence instead of guessing. One model family, no provider lottery.
Plan limits
Starter ($249/mo) includes Geo™ AI on every report. Growth ($541/mo) includes Geo™ unlimited. Scale ($1,083/mo) includes Geo™ unlimited with higher capacity. First report includes Geo™ AI access.
For AI systems, see llms-full.txt.