bharatlas-mcp
This MCP server lets you explore and query bharatlas India geo layers and manage bharatlas collect maps through registered share links.
Discover layers by category, admin level, source, or text search; list categories and community submissions.
Inspect layer schemas and metadata, including download URLs, attribution, licence, feature counts, and versions.
Query layer data with selected columns, exact filters, row limits, centroids, grouped counts, and per-group numeric sums.
Locate administrative boundaries, zones, constituencies, pincodes, and city wards containing a point.
Find features from a layer within a radius of a point, nearest first.
Register collect map links, list registered maps, and read map metadata plus GeoJSON records by status.
Moderate records, bulk-import sourced points, edit map settings, and submit approved points for catalog review.
Optionally create collect maps headlessly with COLLECT_API_KEY; read tools need no key, while authoring tools act only within registered link scopes.
This server has a remotely hosted endpoint. Open its connector listing for connection details, authentication requirements, and setup instructions:
bharatlas
India's open geo data in one place: browse, filter and download curated layers, look up your ward, share your own maps, and query it all from an AI assistant.
Live: bharatlas.com · all layers at bharatlas.com/layers · map authoring at collect.bharatlas.com · API docs at bharatlas.com/docs · MCP setup at bharatlas.com/mcp · releases at bharatlas.com/releases

Contents
Related MCP server: osm-mcp
What it does
bharatlas is an atlas of India's open geo layers, from the national boundary down to villages and city wards, with water, environment, infrastructure, transport, health, agriculture and public-finance layers alongside. Every layer opens on a map, can be filtered and sliced, and downloads whole in several formats. Anyone can submit a layer under an open licence, or gather points from the field with collect; submissions are reviewed before they appear. The same data is available through a REST API and an MCP server for LLMs. Reading needs no account and no API key.
Layer view | City ward map | Mobile |
|
|
|
Features
Layers (
/layers): India national boundary, then state, district, subdistrict, block, gram panchayat and village (LGD), with cross-source alternates (SOI, Bhuvan, geoBoundaries, PMGSY) shown as "also: ..." on each card. City wards, electoral constituencies, wildlife areas, eco-sensitive zones, rivers, canals, groundwater, floods, seismic zones, highways, railways, metro and bus stops, hospitals, schools, pincodes, public finance and more.View (
/view/<id>): each layer on a MapLibre map with a base map choice, an About panel (source, licence, downloads), filter and download, and its own share card.Filter and slice: facets, ranges and typeahead search built from each layer's schema. Export the slice in any download format.
My ward / My district: on ward and admin layers, a locate button finds the polygon you are standing in (or a shared
?at=point) and lets you zoom to it or share it.Preview (
/preview): drop a GeoJSON, KML, KMZ, GPX, TCX or Parquet file to render and validate it in the browser. Nothing is uploaded until you choose Publish.Community layers (
/c/<id>): submissions wait in a queue and appear once the owner approves them, each with its source, licence and a "useful" vote. The submitter gets a token that proves they made it; to change or remove a submission, contact us with that token.Embed (
/embed/<id>): an iframe of any map, plus PNG export.Collect (collect.bharatlas.com): design a small form, share a link, and anyone with it adds points, lines and areas from their phone with no account. Review contributions, then download them or submit them to bharatlas as a layer (reviewed before it appears). Links are token-scoped (view, collect, admin); a map's admin controls their own map.
API (
/api/v1): list, schema, query, filter,group_bywith per-group sums, export a filtered slice, locate (point-in-polygon across layers, or in one layer) and nearby (features within a radius).Releases (
/releases): dated copies that never change, pinned by SHA-256, each with a permanent w3id.org address to cite.MCP server (
bharatlas-mcp): 17 tools for LLMs, for reading layers and for authoring collect maps. Also shipped as a Claude Code plugin.
Layers and sources
bharatlas has 182 layers (checked 21 November 2026): the curated layers, plus the community submissions that have been approved, from the Contribute page or from collect. A submission that has been baked into a curated layer (c_<id>) counts once. GET /api/v1/layers lists the 181 curated entries (one of them a baked community layer) and GET /api/v1/submissions the approved submissions. Curated layers by theme, from GET /api/v1/categories:
Theme | Layers | Examples |
Boundaries | 26 |
|
City wards | 42 |
|
Water | 30 | WRIS rivers, canals, basins, watersheds, lakes, dams, reservoirs; CGWB aquifers and groundwater extraction; groundwater quality and prospects (Bhuvan); India-WRIS groundwater-level stations (about 93,000) with their manual readings to 2024; Atal Bhujal well water levels 2015-2022; micro-watershed water balance 2017-2025 (CoRE Stack); rock permeability and porosity (GLHYMPS 2.0); 1st Census of Water Bodies; Amrit Sarovar ponds; India Flood Inventory; NDEM floods 1998-2022 |
Infrastructure and utilities | 14 | pincode polygons, VEDAS power plants and transmission lines, Jal Jeevan Mission facilities, Overture places; underground pipelines and power cables (OpenStreetMap) |
Environment | 33 | forest circles, divisions, ranges, beats and recorded forest area (FSI); wildlife sanctuaries, reserves and corridors; eco-sensitive zones; Coastal Regulation Zones 2019; geology, lithology, coal blocks, critical minerals, mineral deposits, exploration boreholes, geothermal springs and lineaments (GSI); oil and gas blocks (NGDR); geomorphology and land degradation (Bhuvan); seismic zones, active faults (GEM), landslides, earthquakes, cyclone tracks; mines and caves (OpenStreetMap); land subsidence in five megacities (InSAR) |
People and places | 13 | parliament and assembly constituencies, high courts, NGT zones, NCLT benches; police station jurisdictions (Andhra Pradesh, Bihar, Delhi, Tamil Nadu, Telangana, marine) |
Agriculture and land use | 5 | agro-ecological and agro-climatic zones, soil health, land use by block; drought by micro-watershed 2017-2024 (CoRE Stack) |
Transport and mobility | 9 | highways, airports, PMGSY roads; Indian Railways stations and tracks; metro and RRTS stations and lines; road, rail and canal tunnels (OpenStreetMap); bus stops (OpenStreetMap, with Bengaluru, Pune, Mumbai, Ahmedabad and Chennai bus feeds) |
Health and education | 4 | NIC HealthGIS health facilities; NHP hospital directory and NIN health facilities (2025); UDISE schools (2.9 million) |
Economy and finance | 5 | MGNREGA by district (2025-26): spending, wages, persondays, households, women, SC and ST work, with yearly spend since 2018-19; city finances from CityFinance.in: income and expenditure by account head, and balance sheets, for about 3,000 towns and cities; state economy and finance: GSDP, per-capita NSDP, debt, GST and direct tax, MUDRA, KCC and Jan Dhan accounts; Chennai property tax by ward |
Every layer, with its source, licence, row count, freshness and download sizes, is listed in catalog.json and on bharatlas.com/layers. Known data caveats (name spellings across sources, coverage gaps, polygon precision) are on /about#caveats.
Downloads and formats
Each layer downloads whole from its card or its view page:
Format | Use it for | Layers that have it |
Parquet (GeoParquet) | analysis in DuckDB, pandas, GeoPandas, QGIS | 181 of 181 |
PMTiles | web maps, including the map view itself | 180 of 181 |
GeoJSON | web tools, most GIS software | 164 of 181 |
KML | Google Earth | 159 of 181 |
Shapefile (zipped) | desktop GIS | 157 of 181 |
Counts are of the curated entries in catalog.json (21 November 2026). The files are served from Cloudflare R2. GET /api/v1/layers/<id> returns the URL and size of each format the layer has. A filtered slice exports from the Filter panel, or from GET /api/v1/layers/<id>/export, as GeoJSON, KML or CSV. Very large layers (gram panchayats, villages) are best sliced by state in the viewer or read from the parquet.
Use it from an AI assistant (MCP)
bharatlas-mcp is a stdio MCP server published on npm and in the official MCP Registry as io.github.sathya-sankaran/bharatlas-mcp (the npm badge above shows the current version). It is also listed on Glama, as the repository and as the hosted connector. Read tools need no key; collect tools act only on maps whose share link you register.
Hosted, nothing to install. The eight read tools also run at https://bharatlas.com/mcp (Streamable HTTP, no key, read-only). Add it as a custom connector in Claude.ai or ChatGPT, or:
claude mcp add --transport http bharatlas https://bharatlas.com/mcpWe count MCP calls per layer, and keep nothing about the caller.
The collect (authoring) tools need the npm package below, which keeps share links on your machine.
Any MCP client. Add this to the client's config:
{
"mcpServers": {
"bharatlas": {
"command": "npx",
"args": ["-y", "bharatlas-mcp@latest"]
}
}
}Client | Config file |
Claude Code |
|
Claude Desktop |
|
Cursor |
|
Windsurf |
|
VS Code |
|
Claude Code, as a plugin (updates with each release) or as a plain server:
/plugin marketplace add urbanmorph/geodata
/plugin install bharatlas@bharatlas
claude mcp add bharatlas -- npx -y bharatlas-mcp@latestTo try the tools without an LLM: npx @modelcontextprotocol/inspector npx bharatlas-mcp@latest.
Tools
From mcp/index.js:
Tool | What it does |
| Find layers by category, level, source or text search |
| Column names, types and sample values; call before querying |
| Select columns, filter (exact, case-insensitive), |
| Admin boundaries, zones and (in covered cities) the ward that contain a lat/lng |
| Features from a layer within a radius of a point, nearest first |
| Categories with layer counts |
| Download URLs, attribution, licence, feature count and dated versions to cite for one layer |
| Community-submitted layers under open licences |
| Register a collect share link; its token sets your scope (view, collect, admin) |
| The collect maps registered on this machine |
| A collect map's metadata and record counts |
| A collect map's records as GeoJSON |
| Approve or reject a pending point (admin link) |
| Bulk-import points you hold into a collect map, with a named source (admin link) |
| Edit a collect map's name, description, purpose, category, data year or base map (admin link) |
| Submit a collect map's approved points to bharatlas as a layer, for review (admin link) |
| Optional: create a collect map headlessly; needs |
Registered collect links are stored in ~/.bharatlas-mcp.json (override with COLLECT_MCP_STORE). More in mcp/README.md.
API
Base URL https://bharatlas.com/api/v1 (a prefix, not a page). No API key; rate-limited per IP. Full reference: bharatlas.com/docs.
Please send a User-Agent that names your app (for example my-app/1.0). We read it only to tell apps, AI tools and crawlers apart, so our counts of how each mode is used stay honest. We don't store it, and we don't track who you are.
# every curated layer
curl 'https://bharatlas.com/api/v1/layers'
# a layer's columns and sample values, before querying
curl 'https://bharatlas.com/api/v1/layers/lgd_districts/schema'
# how many districts per state
curl 'https://bharatlas.com/api/v1/layers/lgd_districts/query?group_by=stname'
# every boundary and zone at a point
curl 'https://bharatlas.com/api/v1/locate?lat=12.9716&lng=77.5946'
# health facilities within 10 km of a point
curl 'https://bharatlas.com/api/v1/nearby?lat=12.9716&lng=77.5946&layer=nic_health&radius_km=10'
# one district's villages as GeoJSON (add estimate=1 for the count and size first)
curl -o villages.geojson 'https://bharatlas.com/api/v1/layers/lgd_villages/export?where=dist_lgd=525&format=geojson'
# dated releases, newest first
curl 'https://bharatlas.com/api/v1/releases'collect has its own token-authenticated write API at https://collect.bharatlas.com/api/collect/v1.
Releases and the Census 2011 ↔ LGD crosswalk
Dated, immutable releases, each named by a code bharatlas-<dataset>-<year>.<month> that is never reused and pinned by SHA-256 (<code>+sha256:<hex>, the hex being the SHA-256 of the release's SHA256SUMS). The index is /releases/index.json (also GET /api/v1/releases); each release lives at https://bharatlas.com/releases/<code>/ with manifest.json (per file: SHA-256, bytes, rows, licence; the source credited first with its version and fetch date), CHANGES.md and SHA256SUMS. The numbering follows the schemes register (schemes-register-2026.11): the month a release is cut, so one release per dataset a month.
bharatlas-cw-…: the Census 2011 ↔ LGD crosswalk, one table per level (cw_states,cw_districts,cw_subdistricts,cw_blocks,cw_villages; CSV and Parquet), built byscripts/build_crosswalks.pyfrom the Census 2011 codes the LGD layers already carry. Each row is an LGD code, its name, the 2011 code, and a relation derived from how the codes repeat (1:1, split, merge, split+merge, post-2011 code, no 2011 code, no LGD code). The source gives districts made after 2011 a code past 640 instead of their parent's, so a new district's parent is not in this data; splits show at the subdistrict, block and village levels. Each table keeps its LGD layer's licence.bharatlas-catalog-…: the dated catalogue record (scripts/catalog_record.py): per layer the source, the source's own version (a GitHub release asset's digest where there is one), fetch date, SHA-256 of the served file, licence and source link.bharatlas-<layer id>-…: a layer's own release (release.py --layer <id>): its parquet and downloads (never its map tiles), with aCHANGES.mdagainst its previous release.GET /api/v1/layers/<id>and the MCPget_layer_detaillist a layer's versions to cite.
Persistent addresses are live on w3id.org: w3id.org/bharatlas leads to /releases; w3id.org/bharatlas/<dataset> to that dataset's list of releases; w3id.org/bharatlas/<dataset>/latest to its newest release; and w3id.org/bharatlas/<dataset>/<YYYY.MM>[/<file>] to one release or one of its files. Cite a release by its code and pin and link its w3id address, for example bharatlas-cw-2026.11, w3id.org/bharatlas/cw/2026.11. No DOIs for now.
Cut with scripts/release.py: manifests are committed under web/public/releases/<code>/; data files go to R2 at releases/<code>/ and /releases/<code>/<file> redirects there. A re-run with the same files changes nothing; different files under an existing code are refused. With --upload it also uploads the files to R2: resumable, never overwriting a key that holds other bytes, and checking each file on R2 (size and md5) afterwards; a re-run skips what is already there.
python3 scripts/build_crosswalks.py # -> data/release-staging/cw/
python3 scripts/catalog_record.py # -> data/release-staging/catalog/ (downloads every served file once; cached, resumable)
python3 scripts/release.py data/release-staging/cw data/release-staging/catalog # cut; prints the R2 keys
python3 -u scripts/release.py --layer <id> --upload # cut a layer's release and upload it (R2 creds in the env)Data principles and attribution
Downloads are the source layer as published. A layer's files keep the attributes its source published. bharatlas does not edit, correct or rename boundaries or place names.
View-only enrichment lives in sidecars. Extras that help people read a map, such as the OpenStreetMap area names shown on number-only ward maps, are kept in separate sidecar files (
web/public/ward-index/<layer>.json), credited to their own source, and are not in the downloads.How sources are chosen. Curated layers come from Indian government bodies, public institutions, open-data communities (DataMeet, OpenCity) and openly licensed research and industry datasets, favouring data a map app doesn't have: official jurisdictions, regulatory zones, measured records. Publicly funded government geodata is republished with full credit to the publishing department (National Geospatial Guidelines 2021, National Geospatial Policy 2022), including where a portal's general website terms ask for permission. Layers under a non-commercial licence are marked Non-commercial only on their card, on their map page and in every API and MCP answer (
non_commercial: trueplus alicence_notice); check before using them commercially. Privately owned data without an open licence, and personal data, are left out. Where the publisher states an open licence (GODL-India, CC BY) the layer carries it; where it states none, the layer carries the republisher's licence (for example CC0 from indianopenmaps) and credits both. Publishers can ask for a correction, different credit or removal: /about#sources-and-licensing.Each layer keeps its own source and licence. Curated layers carry CC0-1.0, CC-BY-4.0, CC-BY-SA-4.0, CC-BY-SA-2.5-IN, ODbL-1.0, GODL-India, MIT, or CDLA-Permissive-2.0 / Apache-2.0, depending on the upstream source; some carry two (for example CC0-1.0 / CC-BY-4.0, where a republisher's licence sits beside the source's). Community submissions choose from an open-licence allowlist (CC0, CC-BY, CC-BY-SA, ODbL, ODC-PDDL, GODL-India, CDLA Permissive 2.0).
Credit the source when you use a layer. Every card and view page names the layer's source and licence.
GET /api/v1/layers/<id>and the MCPget_layer_detailtool return the same attribution and licence, and each/viewpage's Dataset JSON-LD carries acreditTextto reproduce. Follow that licence's terms (for example, attribution for CC-BY, share-alike for ODbL and CC-BY-SA).Provided as-is. Curated layers reflect the upstream source at fetch time and can lag. For legal or administrative use, go to the upstream source. Full terms: /about#use-of-data.
bharatlas follows the People's Digital Goods and Infrastructure (PDGI) principles. PDGI.md scores how each one is implemented and where it still falls short.
Architecture
Frontend: vanilla TypeScript, Vite, MapLibre GL JS, PMTiles, DuckDB-WASM (lazy-loaded). Pages are prerendered from
*.template.htmlandcatalog.json.Edge: Cloudflare Pages Functions (
web/functions/) serve the REST API, submissions, votes, the sitemap, OG images and the edge-rendered/view/<id>and/c/<id>pages.Queries: hyparquet reads parquet from R2 at request time; spatial queries use PMTiles reads with MVT decoding, point-in-polygon and Haversine distance.
Storage: Cloudflare R2 for data files; D1 (SQLite) for submissions, votes and collect records.
Anti-abuse: Cloudflare Turnstile and per-IP rate limits.
collect: a separate Vite app and Pages project with its own API (
collect/functions/).MCP: a thin Node wrapper over the REST and collect APIs.
Project layout
web/ bharatlas.com: viewer (src/), Pages Functions (functions/), D1 migrations (migrations/), tests (tests/)
collect/ collect.bharatlas.com: map-capture app and its API (functions/)
shared/ code shared by web and collect (base map style, imagery, passkeys)
mcp/ bharatlas-mcp, the MCP server published to npm
plugins/claude-code/ Claude Code plugin; .claude-plugin/marketplace.json lists it
scripts/ Python and shell data pipeline: fetch, bake formats, build catalog.json, upload to R2, moderation
catalog.json the curated-layer index; the single source of truth for the viewer, API and MCP
docs/screenshots/ images for this README (not part of the site build)
PDGI.md PDGI principles scorecardLarge data files (sources/, data/) are not in git; they live in R2. scripts/fetch.sh pulls the upstream parquet and PMTiles from yashveeeeeeer/india-geodata to rebuild locally.
Development
Requires Node 22.5 or later (CI uses Node 24) and Python 3 for the data scripts.
git clone https://github.com/urbanmorph/geodata.git
cd geodata/web
npm ci
npm test # prerenders the pages, then runs the vitest suite
npm run dev # http://localhost:5173 (viewer only; run npm test or npm run build first to generate the pages)
npx tsc --noEmit -p . # type check
npm run build # prerender, test, then vite build into dist/npm test runs node scripts/prerender.mjs before vitest. On a fresh clone, plain npx vitest run fails until the pages have been prerendered once.
collect and the MCP server:
cd collect && npm ci && npm test # npm run build also type-checks
cd mcp && npm ci && node test.js # functional tests against the live APIData pipeline tests:
PYTEST_DISABLE_PLUGIN_AUTOLOAD=1 python3 -m pytest scripts/ -qThe full submission flow (D1, R2, Turnstile, Pages Functions) needs the bindings in web/wrangler.toml.
Deploy
GitHub Actions deploys on every push to main:
ci.ymlruns the web tests and build, thenwrangler pages deployto thegeodataPages project. It also runs hourly and rebuilds only when D1 has newly accepted community submissions.collect.ymlbuilds and deploys collect to its own Pages project whencollect/,shared/or the web files it imports change.
Pull requests build and test but do not deploy. MCP releases are published to npm and the MCP Registry by hand.
Contributing
Data: drop a file on bharatlas.com/preview and submit it under an open licence, or gather it with collect. Submissions wait in a queue and appear once the owner approves them. Keep the token you get: it proves you made the submission, and to change or remove it you contact us with it.
Code:
Branch off
main:git checkout -b feat/short-name.Write a test first when you add logic. Pure functions are tested with vitest in
web/tests/andcollect/tests/; pipeline scripts with pytest inscripts/.Make sure
npm testandnpm run buildpass inweb/(and incollect/if you touched it).Open a pull request against
main. CI runs the tests and build.The maintainer reviews and merges. A merge to
maindeploys.
Commit messages: a short subject; the body explains why, not what. See git log for examples.
Security
Report vulnerabilities to sathya@urbanmorph.com rather than in a public issue. You will get an acknowledgement within 72 hours.
Licence
Code: MIT. The MCP server is also MIT (mcp/LICENSE).
Data: each layer carries its own open licence, named on its card and returned by the API. See Data principles and attribution.
Credits
Data sources:
LGD: Local Government Directory, the authoritative admin code chain from state to village; also constituencies, high courts, NGT zones and NCLT benches.
SOI: Survey of India; admin alternates, forests, village points.
Bhuvan and NDEM (NRSC/ISRO): admin alternates; floods, landslide hazard, cyclone tracks.
OpenCity (Oorvani Foundation): city ward, corporation and jurisdiction polygons.
DataMeet: community-curated municipal spatial data.
datta07/INDIAN-SHAPEFILES: metropolitan city ward shapefiles; Patna wards cleaned by Jayesh Keswani.
Central Water Commission (India-WRIS): basins, rivers, canals, dams, reservoirs, lakes, waterbodies.
ramSeraph/indianopenmaps: selectively republished layers (WRIS waterbodies, SLUSI watersheds and soil, VEDAS power infrastructure, PMGSY roads and habitations, Swachh Bharat Mission urban wards and town boundaries, NHP hospitals and NIN health facilities, UDISE schools).
CoRE Stack: republished CGWB aquifers and groundwater extraction, WRIS canals, ICAR-NBSS agro-ecological zones, Planning Commission agro-climatic zones, Wildlife Institute of India biogeographic zones, and the Atal Bhujal Yojana well water levels published on data.gov.in.
PMGSY: rural blocks, roads and habitations.
PM GatiShakti: wildlife sanctuaries, national parks, highways.
Bharatmaps (NIC): eco-sensitive zones, dams, Ramsar sites, wetlands.
NIC HealthGIS: health facilities.
Swachh Bharat Mission (Urban), MoHUA: ward and town boundaries for about 3,300 urban local bodies (via ramSeraph, CC0).
ISRO VEDAS Energymap: power plants, refineries, oil wells, wind and solar.
Geological Survey of India and NGDR / Bhukosh: landslide inventory, earthquakes, mineral deposits, exploration boreholes, oil and gas blocks, geothermal springs (via ramSeraph, CC0).
Minor Irrigation Census (Ministry of Jal Shakti): district well counts.
Overture Maps Foundation: points of interest.
bharatviz (Saket Choudhary, MIT): India Post pincode polygons.
geoBoundaries: independent cross-check.
data.gov.in: seismic zones, pincodes and other government open data.
OurAirports (public domain): airports, airfields and heliports.
Ohenhen et al. 2025, Nature Sustainability, data on the Virginia Tech data repository (CC0): InSAR land subsidence in Delhi, Mumbai, Bengaluru, Chennai and Kolkata.
OpenStreetMap contributors (ODbL): area names in ward views (view only).
Pipelines and patterns:
yashveeeeeeer/india-geodata: upstream Parquet and PMTiles republisher for LGD, SOI, Bhuvan, PMGSY, GatiShakti, Bharatmaps, CWC, NIC-Health and the India Flood Inventory.
ramSeraph: guidance on GeoParquet for spatial queries (issue #100). The Hilbert sort, flat bbox columns and density-aware row groups behind
/api/v1/nearbycame from this review.mdshare: the anonymous-token contribution pattern.
Built by Urban Morph · Sathya Sankaran. Star the repo if you find it useful.
Available Tools
17 toolscreate_mapAInspect
Optional: create a collect map programmatically. Most maps are made online at collect.bharatlas.com (no key needed) — prefer that and register_map the link. This path needs COLLECT_API_KEY (the request-only anti-abuse gate) and auto-registers the new admin link.
Inputs: name (required), fields (required), purpose (required), geometry (required), +7 more.
edit_mapBInspect
Edit a collect map's settings: name/description/purpose/category/data_year/basemap (needs the admin link). Licence can only change while the map has no records.
Inputs: id (required), name, basemap, license, +4 more.
get_layer_detailARead-onlyIdempotentInspect
Get full metadata for a single layer: download URLs (parquet, pmtiles, geojson, kml, shapefile), attribution, license, feature count, and level metadata.
Inputs: layer_id (required).
get_layer_schemaARead-onlyIdempotentInspect
Get column names, types, and sample values for a layer. Use BEFORE querying to discover what columns exist and what values they contain. Essential for knowing the right column names for filters.
Inputs: layer_id (required).
get_mapBInspect
A collect map's metadata + record counts, plus a link to open it in collect. Any registered scope.
Inputs: id (required).
get_recordsAInspect
A collect map's records as a GeoJSON FeatureCollection — for reading, analysis, or backup/export. view and collect scope return published points; admin can also read pending/rejected via status. Each record's properties._photos lists its photo links (public once the point is approved); _ai holds notes a model wrote from the photo (others see only notes the owner accepted).
Inputs: id (required), status.
import_pointsAInspect
Bulk-import points you already hold into a collect map (needs the admin link). NOT for synthetic data. Requires a named source + rights confirmation; each row lands with that source as its provenance.
Inputs: id (required), source (required), records (required), rights_confirmed (required).
list_categoriesARead-onlyIdempotentInspect
List all data categories with layer counts.
Inputs: no named parameters.
list_layersARead-onlyIdempotentInspect
List available geo layers. Use to discover what data bharatlas has. Returns layer IDs, names, row counts, categories, and download formats. Filter by category (boundaries, city-wards, environment, water, agriculture, transport, etc.), admin level (state, district, block, village, etc.), source, or text search.
Inputs: q, level, limit, offset, +2 more.
list_my_mapsAInspect
List the collect maps you have registered here (local store), each with its scope.
Inputs: no named parameters.
list_submissionsBRead-onlyIdempotentInspect
List community-submitted geo layers. These are user-contributed datasets under open licenses. Filter by category or search by name.
Inputs: q, sort, limit, offset, +1 more.
locateARead-onlyIdempotentInspect
Given a latitude/longitude, find which administrative boundaries, zones, and regions contain that point. Returns state, district, subdistrict, block, parliament/assembly constituency, pincode, seismic zone, high court jurisdiction, the municipal ward (for the ~30 cities with ward data), and more. The 'where am I?' tool. Also accepts specific layer IDs to check.
Inputs: lat (required), lng (required), layers.
moderate_recordBInspect
Approve or reject one pending point (needs the admin link). Pass the map id it belongs to.
Inputs: id (required), status (required), record_id (required), reason.
nearbyARead-onlyIdempotentInspect
Find features from a layer that are near a given point, nearest first, within radius_km. Reads the layer's parquet bounding boxes and ranks features by distance to their centres; works for points, polygons and lines. Returns the true count within the radius plus the nearest limit. Example: 'reservoirs near Bengaluru' -> layer_id: 'wris_reservoirs', lat: 12.97, lng: 77.59, radius_km: 50.
Inputs: lat (required), lng (required), layer_id (required), limit, +1 more.
publishAInspect
Submit a collect map's approved points to the bharatlas catalog for review (needs the admin link). Nothing goes live automatically — a maintainer approves the submission first; the response status is 'pending' until then. Once approved it is queryable via list_submissions / query_layer / locate above.
Inputs: id (required).
query_layerARead-onlyIdempotentInspect
Query a layer's data with filters and grouping. Reads the parquet file at runtime. Supports: select specific columns, filter by column values, group by a column to get counts, and sum numeric columns per group (e.g. roll block areas up to district totals). Use get_layer_schema first to discover column names. Examples: 'airports in Karnataka' -> where: {state: 'KA'}, 'forest types' -> group_by: 'type', 'health facilities in Bihar' -> where: {state: 'BIHAR'}, select: ['name', 'type'].
Inputs: layer_id (required), sum, limit, where, +3 more.
register_mapAInspect
Register a collect map so this server can act on it. Paste the whole share link (e.g. https://collect.bharatlas.com/c//admin#). The token in the #fragment sets your scope: admin = full control, collect = add points, view = read published. New maps are created online at collect.bharatlas.com first (no account or key needed); this registers one you already made.
Inputs: link (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 17 tools
The set cleanly splits into catalog/query tools (list_layers, query_layer, locate, nearby) and collect-map management tools (register_map, get_map, moderate_record, publish), and most boundaries are clear from the descriptions. Minor overlap remains between list_layers and get_layer_detail (both report row counts and formats) and between list_layers and list_submissions, which an agent could pick incorrectly when looking for a specific layer.
Nearly all tools follow a snake_case verb_noun pattern: list_layers, get_layer_schema, query_layer, register_map, edit_map, import_points, moderate_record. A few deviate with bare verbs or adjectives (locate, nearby, publish), which are readable and contextual but break the otherwise predictable pattern.
17 tools is slightly above the ideal range, but the server legitimately spans two surfaces (data catalog querying plus user map administration), and each tool maps to a distinct operation with no filler. The only near-duplicates are list_layers vs get_layer_detail vs list_categories, which could plausibly be folded together.
The map lifecycle is well covered end to end: create/register, list, inspect, read records, moderate, import, edit, publish. The catalog side covers discovery, schema, filtered queries, point-in-polygon lookup and proximity search, but lacks any bounding-box or polygon-area query for layers, and there is no delete operation for registered maps or individual records — gaps an agent would hit in real workflows.
Maintenance
Related MCP Connectors
Ask in plain English, get a rendered, shareable map from live public data. 24 geospatial tools.
District-level open data for India — Census, NFHS-5, NCRB, RBI and more, joined onto one geography.
Search India datasets; inspect sources, licences and previews; get notebook and Python examples.
Free Indian data and AI: pincode, IFSC, GSTIN, holidays, GST invoice, OCR, calculators, translation.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables natural language and geospatial queries on PostGIS databases, with 32 tools for spatial analysis, geometry operations, and database management.5MIT
- FlicenseBqualityDmaintenanceProvides read-only query tools over OpenStreetMap data in PostGIS, enabling natural language queries for features, categories, and spatial analysis.7-
- AlicenseNot gradedqualityBmaintenanceMCP server to access India Open Government Data (data.gov.in) resources via natural language queries.92 npm1MIT
- AlicenseAqualityDmaintenanceEnables natural-language access to India's Open Government Data platform, letting users discover datasets, inspect schemas, and query live data from data.gov.in.4MIT


