Find the best spot,
with evidence.
Turn a plain-language question — English or Vietnamese — into a site-selection analysis: a 10-minute walking isochrone around each candidate, competitors and demand counted inside it, drive times from real origins, and a published ranking formula.
>_ Terminal
git clone https://github.com/ducvuathuduc/sitescout-agent.git
cd sitescout-agent
npm ci
npm run build
export GEOAPIFY_API_KEY=your_key
node dist/cli.js scout "Where should I open a specialty coffee shop in Thảo Điền?" --id my-coffeeclaude mcp add sitescout -e GEOAPIFY_API_KEY=YOUR_KEY -- node "$PWD/dist/mcp-server.js"GEOAPIFY_API_KEY=YOUR_KEY npm run ask -- --question thaodien-coffeeSet up your coding agent
Paste the prompt into Claude Code, Cursor or Claude Desktop. It points the agent at the machine-readable index and then at the guides.
Read https://sitescout-docs.pages.dev/llms.txt, then follow https://sitescout-docs.pages.dev/docs. Use SiteScout to answer site-selection questions for Ho Chi Minh City and cite every number with its [Tn] reference.Explore the docs
MCP usageFive tools that run a whole analysis from your AI assistant.Demo proxyHow the Worker keeps the key secret and shares one daily budget.Credits and limitsHow each Geoapify call is priced, and the caps the demo enforces.How it worksPlan, geocode, isolines, scoring — and where each step lives.Recorded sessionsReal agent runs, including the failures that shaped the tools.Benchmarks55 HCMC address queries scored against Wikidata coordinates.