{"slug": "placeroot-keyless-map-tools-for-ai-agents-places-geocode-routing", "title": "PlaceRoot – keyless map tools for AI agents (places, geocode, routing)", "summary": "PlaceRoot, an independent project by vibemapper, released an MCP server that grounds AI agents in open map data from Overture Maps, offering 42 tools for spatial questions such as nearby places, geocoding, and routing without requiring an API key or signup. The server reads Overture's public data directly, supports real street-graph routing and isochrones, and can be run via PyPI or npm, with integrations for Claude Code, Claude Desktop, and Cursor.", "body_md": "[Quick start](#quick-start)\n·\n[Tools](#what-it-can-do)\n·\n[Reference](https://github.com/chuofringer/placeroot/blob/main/docs/REFERENCE.md)\n·\n[Benchmarks](https://github.com/chuofringer/placeroot/blob/main/docs/benchmarks.md)\n·\n[Website](https://placeroot.dev)\n\n**PlaceRoot grounds AI agents in open map data.** It's an MCP server that answers spatial questions — what's nearby, what's in this neighborhood, how do I get there — from [Overture Maps](https://overturemaps.org) open data. No API key, no signup, no vendor platform.\n\nAn independent project by [vibemapper](https://vibemapper.dev).\n\n- 🎯\n**Answers, not data dumps.** Every tool returns compact, ranked results sized for an agent's context window — never a raw GeoJSON dump. - 🗺️\n**Real routing, zero keys.**`route`\n\nand`isochrone`\n\nwalk an actual street graph built from Overture's transportation segments — not a straight-line guess — anywhere on Earth. - 🚶\n**Reachability-filtered search.**`find_places(..., within={\"minutes\": 15, \"mode\": \"walk\"})`\n\nkeeps only results truly inside the street-graph walk/cycle/driveshed — not a radius that guesses at it, and not a second call to intersect a polygon yourself. - 🏙️\n**Rich, filterable place data.** Category, brand, confidence, operating status, contactability — sourced from Overture's open dataset (contributed by Meta, Uber, TomTom, and others). - 📐\n**Boundary-accurate.** Search inside a named place's real administrative polygon, not a guessed radius circle. - ⚡\n**Zero setup.** Reads Overture's public data directly — no key, no database, nothing to install beyond the server itself.\n\nRun it straight from PyPI or npm — no install step:\n\n```\nuvx placeroot             # stdio MCP server\nuvx placeroot --http      # HTTP endpoint at http://127.0.0.1:8321/mcp\n```\n\n**Add to Claude Code**\n\n```\nclaude mcp add placeroot -- uvx placeroot\n```\n\n**Add to Claude Desktop**\n\nAdd to your `claude_desktop_config.json`\n\n:\n\n```\n{\n  \"mcpServers\": {\n    \"placeroot\": {\n      \"command\": \"uvx\",\n      \"args\": [\"placeroot\"]\n    }\n  }\n}\n```\n\n**Add to Cursor / other MCP clients**\n\nSame config as above wherever your client keeps its MCP server list. Prefer npm? Use `\"command\": \"npx\", \"args\": [\"placeroot\"]`\n\n.\n\n```\n{\n  \"mcpServers\": {\n    \"placeroot\": {\n      \"command\": \"npx\",\n      \"args\": [\"placeroot\"]\n    }\n  }\n}\n```\n\nThen ask your agent something spatial:\n\n\"What's around downtown Palo Alto?\" · \"Compare these two neighborhoods for a bike shop.\" · \"Plan my errands: pharmacy, hardware store, post office.\"\n\n**42 tools**, grouped into four families — every answer fits in a couple of thousand tokens:\n\n| Family | Tools | Answers questions like |\n|---|---|---|\n🔍 Search & identify |\n`find_places` , `find_near` , `geocode` , `reverse_geocode` , `place_details` , `search_categories` , `warmup_city` , … |\nWhat cafés are near this point? Coffee near the Eiffel Tower? |\n📊 Area analysis |\n`summarize_area` , `compare_areas` , `neighborhood_verdict` , `summarize_buildings` , `land_use_at` , … |\nWhat's in this neighborhood, should I live here, and how does it differ from that one? |\n🚴 Routing |\n`route` , `from_to` , `isochrone` , `optimize_route` , `places_along_route` , `distance_matrix` , `travel_time_matrix` |\nHow far to walk from A to B? What's reachable in 15 minutes? Best order for 6 stops? |\n🗺️ Geometry & maps |\n`render_map` , `simplify_geometry` |\nShow me this result as an interactive map |\n\nIt also ships seven **workflow prompts** (site selection, neighborhood comparison, errand planning, should I live here, get to know my city, verify listing claims, plan area visit) and three attachable **resources** — and a `PLACEROOT_TOOLS`\n\nsetting to load only the tool profiles you need, cutting schema overhead by up to 95%.\n\n📚 [Full tool catalog, prompts, resources & configuration → docs/REFERENCE.md](/chuofringer/placeroot/blob/main/docs/REFERENCE.md)\n\nNote\n\nOpen data has honest limits: no live traffic, no opening hours, no ratings or photos — [what PlaceRoot deliberately doesn't do](/chuofringer/placeroot/blob/main/docs/REFERENCE.md#what-it-deliberately-does-not-do). For everything else about where things are and what's reachable from them, it answers without a key.\n\nIt's the only keyless MCP server doing real graph routing over global open map data — with every tool declaring proper [MCP annotations](/chuofringer/placeroot/blob/main/docs/REFERENCE.md#tool-annotations-and-listing-cache-hints) so clients know which calls are read-only before prompting you. Stable [GERS ids](https://docs.overturemaps.org/gers/) let agents hold onto places across turns; local caching makes repeat queries answer in milliseconds and keeps working offline; and the whole thing is [self-hostable end to end](/chuofringer/placeroot/blob/main/docs/MIRROR.md).\n\nHow it stacks up against Mapbox MCP and Google Maps MCP: [head-to-head benchmarks](/chuofringer/placeroot/blob/main/docs/benchmarks-vs.md) · [token-efficiency numbers](/chuofringer/placeroot/blob/main/docs/benchmarks.md).\n\nOverture's places theme is derived from business listings, which makes it strong on businesses and thin on the places a family goes on a Saturday — playgrounds, neighbourhood parks, dog parks, beaches. Those features aren't missing from Overture, though; they're in a different theme. Overture's `base`\n\ntheme is a direct conflation of OpenStreetMap, and PlaceRoot already queries it for `land_use_at`\n\nand `infrastructure_at`\n\n. The places tools read it too, by default.\n\nNothing is downloaded, built, or hosted — it's one more live scan of the same public Overture release, and it roughly **2.5x** es playground coverage (1,552 vs 674 across New York City in release `2026-07-22.0`\n\n, with 1,013 of them more than 150 m from any places-theme playground). Every places tool answers from both at once, with no other change: same tools, same response shape, same category filters.\n\nThe cost is a second dataset scan per places query (cached like everything else), and these rows carry no `confidence`\n\nor `operating_status`\n\nand are often unnamed — an unnamed playground comes back with `name: null`\n\nrather than being dropped. If you'd rather have the latency than the coverage:\n\n```\nexport PLACEROOT_RECREATION_LAYER=0\n```\n\n`data_version`\n\nreports the layer whenever it's active. Full details, including why live Overpass queries and raw OSM Parquet were measured and rejected: [docs/RECREATION.md](/chuofringer/placeroot/blob/main/docs/RECREATION.md).\n\n```\nuv sync           # install dev dependencies\nuv run pytest     # offline test suite\nuv run ruff check .\n```\n\nSee [CONTRIBUTING.md](/chuofringer/placeroot/blob/main/CONTRIBUTING.md) for dev setup, design rules, and how to propose a tool. Other docs: [CHANGELOG](/chuofringer/placeroot/blob/main/CHANGELOG.md) · [SECURITY](/chuofringer/placeroot/blob/main/SECURITY.md) · [publishing](/chuofringer/placeroot/blob/main/docs/PUBLISHING.md) · [website](/chuofringer/placeroot/blob/main/docs/WEBSITE.md) · [running a data mirror](/chuofringer/placeroot/blob/main/docs/MIRROR.md) · [the recreation layer](/chuofringer/placeroot/blob/main/docs/RECREATION.md)\n\n`hello@placeroot.dev`\n\nDeveloped by [vibemapper](https://vibemapper.dev).\n\nPlaceRoot runs on your machine — no account, no API key, no sign-up. It has no telemetry and sends nothing to us; the only network traffic is your own queries going straight to Overture Maps' public data on AWS S3 (or a mirror you configure). Full details: [placeroot.dev/privacy.html](https://placeroot.dev/privacy.html).\n\nThe code is [MIT](/chuofringer/placeroot/blob/main/LICENSE). The data it queries is the Overture Maps public release, licensed per theme — places under CDLA-Permissive-2.0; the OSM-derived themes (divisions, transportation, base) under ODbL, which asks for attribution on anything user-facing you build from them:\n\n© Overture Maps Foundation · © OpenStreetMap contributors (\n\n[ODbL])\n\nPer-theme obligations: [docs/DATA-LICENSE.md](/chuofringer/placeroot/blob/main/docs/DATA-LICENSE.md).\n\nmcp-name: io.github.chuofringer/placeroot", "url": "https://wpnews.pro/news/placeroot-keyless-map-tools-for-ai-agents-places-geocode-routing", "canonical_source": "https://github.com/chuofringer/placeroot", "published_at": "2026-08-26 15:35:57+00:00", "updated_at": "2026-08-26 15:46:01.979854+00:00", "lang": "en", "topics": ["ai-tools", "ai-agents", "developer-tools"], "entities": ["PlaceRoot", "vibemapper", "Overture Maps", "Claude Code", "Claude Desktop", "Cursor"], "alternates": {"html": "https://wpnews.pro/news/placeroot-keyless-map-tools-for-ai-agents-places-geocode-routing", "markdown": "https://wpnews.pro/news/placeroot-keyless-map-tools-for-ai-agents-places-geocode-routing.md", "text": "https://wpnews.pro/news/placeroot-keyless-map-tools-for-ai-agents-places-geocode-routing.txt", "jsonld": "https://wpnews.pro/news/placeroot-keyless-map-tools-for-ai-agents-places-geocode-routing.jsonld"}}