# LatLong Maps (remote · geocp.latlong.ai)

Turn CSV or Excel data into styled choropleth and categorical maps of India from plain language.

- Trust score: 57/100 (low)
- Change this week: +2
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-10-02

## Components

- remote · `geocp.latlong.ai`: 57/100 (this document), [markdown](https://verifymcp.io/servers/ai-latlong-map-mcp/geocp.md), [page](https://verifymcp.io/servers/ai-latlong-map-mcp/geocp)

## Channel facts

- Endpoint: `https://geocp.latlong.ai/mcp/`
- Transports: `streamable-http`
- Auth: `required`
- Version: `1.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-10-02.

- **Endpoint Security**: 51/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation not fully verified: no authorisation is required to call this server, and 5 tool(s) never declared a destructiveHint. The MCP spec treats an absent hint as destructive by default, so we cannot call this surface safe.
  - HTTPS enforcement could not be verified: the plaintext port redirected somewhere other than https, so enforcement is unproven.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 48/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 5223 tokens (~1044/item across 5 items; 5 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 15/100
  - Stability check failed: schema churn in the 30 days we've observed: 5 tool removals, 0 breaking changes, 0 auth/transport breaks, 4 additions.
- **Tool Coverage**: 71/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 0% of tool parameters carry a description.
  - Structured output schemas are declared (40% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 5 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 6 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the LatLong Maps MCP server?

LatLong Maps is a hosted endpoint at https://geocp.latlong.ai/mcp/, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http ai-latlong-map-mcp 'https://geocp.latlong.ai/mcp/'
```

### Cursor

```json
{
  "mcpServers": {
    "ai-latlong-map-mcp": {
      "url": "https://geocp.latlong.ai/mcp/"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "ai-latlong-map-mcp": {
      "type": "http",
      "url": "https://geocp.latlong.ai/mcp/"
    }
  }
}
```

### Codex

```toml
[mcp_servers.ai-latlong-map-mcp]
url = "https://geocp.latlong.ai/mcp/"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "ai-latlong-map-mcp": {
      "type": "remote",
      "url": "https://geocp.latlong.ai/mcp/",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add ai-latlong-map-mcp --url 'https://geocp.latlong.ai/mcp/' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  ai-latlong-map-mcp:
    url: "https://geocp.latlong.ai/mcp/"
```

### Netclaw

```json
{
  "McpServers": {
    "ai-latlong-map-mcp": {
      "Transport": "http",
      "Url": "https://geocp.latlong.ai/mcp/"
    }
  }
}
```

### Vellum

```bash
assistant mcp add ai-latlong-map-mcp -t streamable-http -u 'https://geocp.latlong.ai/mcp/'
```

### Other

```json
{
  "mcpServers": {
    "ai-latlong-map-mcp": {
      "type": "http",
      "url": "https://geocp.latlong.ai/mcp/"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-10-02 (score 57, +1)

No change was recorded against any check on this day. Stability & Change Management went from 11 to 15.

### 2026-10-01 (score 56, 0)

- [security] The server rewrote its instructions, which are the text every model session reads

### 2026-09-30 (score 56, +1)

No change was recorded against any check on this day. Stability & Change Management went from 5 to 8.

### 2026-09-29 (score 55, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [functional regression] Schema quality: 939 → 1044

### 2026-09-28 (score 55, 0)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-09-26 (score 55, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “question_to_map” rewrote its description, which is the text the model reads
- [functional regression] Tool coverage: 67% → 40%
- [functional improvement] Schema quality: 1271 → 939
- [functional] New tool “geocode”
- [functional] New tool “reverse_geocode”

### 2026-09-25 (score 55, +10)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-09-24 (score 45, −9)

- [security regression] Judged manipulation: pass → unverified
- [security regression] A breaking change shipped without a version bump: still 1.0.0
- [security regression] Tool “create_upload” was removed
- [security regression] Tool “process_upload” was removed
- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “get_capabilities” rewrote its description, which is the text the model reads
- [security] Tool “question_to_map” rewrote its description, which is the text the model reads
- [functional regression] Tool coverage: 100% → 67%
- [functional regression] Schema quality: good → unverified
- [functional improvement] Schema quality: 4775 → 3815
- [functional] New tool “upload_file”
- [cosmetic] “question_to_map” added an optional parameter “file_key”
- [cosmetic] “question_to_map” added an optional parameter “sheet”
- [cosmetic] “question_to_map” dropped the optional parameter “csv”
- [cosmetic] “question_to_map” dropped the optional parameter “file_path”
- [cosmetic] “question_to_map” dropped the optional parameter “upload_id”

## MCP tools (5)

### `geocode` (~246 tokens)

Geocode addresses from an uploaded CSV or XLSX file into latitude/longitude. Pass a file_key from upload_file.

Column selection: if you know it, pass address_column (or address_columns for several). Otherwise the server asks the geo-match service to detect it. When the answer is not unambiguous — several address columns, a weak match, or only structured columns like District/State/PIN that must be combined — the response is status=clarification_required with candidates and ready-to-fire guidance actions. Present the choice to the user and re-call with their answer; never invent a column name.

The result overwrites the same file_key as a new version, plus a small preview, not the full rows. Pass that file_key to question_to_map to map the coordinates, or hand the user file_url (a signed download link, when present) to fetch the file directly.

Input parameters:

- `address_column` (string)
- `address_columns` (null|array)
- `compose_columns` (null|array)
- `file_key` (string)
- `force_new` (boolean)
- `job_id` (string)
- `sheet` (string)

### `get_capabilities` (~55 tokens)

Return the server's capability matrix: supported map types, geo types, output formats, classification methods, and the list of tools available in this process (which varies by configuration). Call this first to discover what the server can do.

Output parameters:

- `auto_level_detection` (boolean)
- `available_indices` (null|array)
- `available_tools` (null|array)
- `classification_methods` (null|array)
- `default_output` (string)
- `geo_detection` (boolean)
- `geo_levels` (null|array)
- `guidance` (null|object)
- `index_modes` (null|array)
- `map_types` (null|array)
- `output_formats` (null|array)
- `png_supported` (boolean)
- `stateless_only` (boolean)
- `svg_default` (boolean)

### `question_to_map` (~649 tokens)

Render a geographic map from your data. Call upload_file first to ingest a CSV/XLSX file, then pass the returned file_key: {"question": "show total by pincode", "file_key": "mcp-uploads/2026/01/15/<uuid>.csv"}. The file_key is reusable — for additional questions on the same file, pass the same file_key again without re-uploading.

Alternatively, pass rows directly as 'data' (e.g. from the upload_file preview): {"question": "show total by pincode", "data": [<rows>]}

geo_column and value_column are preferred if easily identifiable from the column names, but not required — the server auto-detects them using a geo-match service that matches your data against known Indian geographies (states, districts, ACs, PCs, pincodes, RTOs, cities).
Hints: geo_column usually contains place names (state, district, city, pincode, constituency) or numeric codes (pincode=560049); value_column usually contains the metric to plot (sales, count, total, percentage) or a category (party affiliation, status, type).
If value_column is 'total' and doesn't exist in data, all numeric columns are summed automatically. String value columns render as categorical maps (each distinct value gets its own color, e.g. party affiliation). If the server cannot determine which columns to use, it returns a 'clarification_required' response with suggested columns and guidance — pick a geo_column and value_column from the suggestions and re-call.
Supported geo types: state, district, pincode, rto, city, ac, pc.

Points mode: if your file has latitude/longitude columns (e.g. from a prior geocoding step, or coordinates you already had) and your question asks to plot/pin/mark locations ("plot my stores", "show me the outlet locations", "mark these points"), the server detects the coordinate columns automatically and renders markers instead of shaded regions — no geo_column/value_column needed. The boundary lines drawn beneath the markers and how far the map zooms are chosen automatically from where the points ac…

Input parameters:

- `data` (null|array)
- `detected_geo` (null|object)
- `enrich_with` (null|object)
- `file_key` (string)
- `geo_column` (string)
- `geo_level` (string)
- `map_type` (string)
- `output_format` (string)
- `question` (string, required)
- `render_params` (object)
- `sheet` (string)
- `state_hint` (null|array)
- `style` (object)
- `table` (string)
- `value_column` (string)

Output parameters:

- `aggregation_applied` (null|object)
- `clarification` (null|object)
- `computed_data` (null|object)
- `confidence` (number)
- `guidance` (null|object)
- `level_resolution` (null|object)
- `map` (null|object)
- `mode` (string)
- `points_plotted` (integer)
- `points_rejected` (integer)
- `rejected_rows` (null|array)
- `resolution_method` (string)
- `resolved_params` (null|object)
- `usage` (null|object)
- `warnings` (null|array)

### `reverse_geocode` (~229 tokens)

Turn latitude/longitude coordinates in an uploaded CSV or XLSX into addresses. Pass a file_key from upload_file.

Column selection: pass latitude_column and longitude_column, or latlong_column for a single combined column, or coordinate_pairs for several sets. Otherwise the server detects them. When the answer is not unambiguous — several coordinate columns, a weak match, coordinates outside the supported region, or coarse whole-number values — the response is status=clarification_required. Present the choice to the user and re-call with their answer.

Results are written back into the uploaded file: the response carries file_key, a signed file_url, and a small preview. Each pair adds address, pincode, landmark and status columns.

Input parameters:

- `accept_low_precision` (boolean)
- `coordinate_pairs` (null|array)
- `file_key` (string)
- `force_new` (boolean)
- `job_id` (string)
- `latitude_column` (string)
- `latlong_column` (string)
- `longitude_column` (string)
- `sheet` (string)

### `upload_file` (~325 tokens)

Upload a CSV or XLSX file for server-side ingestion under policy (auth, size cap, malware scan, injected-intent scan) and receive a file_key. This is the first step before calling question_to_map with your own data.

Step 1: Call upload_file with file_name, file_size, and question. The response contains a signed upload_url and a curl command — run the curl command to POST the file bytes. The curl response contains file_key, columns, a row preview, and optionally 'intent' (geo-match service detected geo_column, geo_level, value_column, map_type, confidence) and 'guidance' (a ready-to-fire question_to_map action with all detected parameters pre-filled). Forward the intent fields into question_to_map to skip auto-detection and render immediately.

For multi-sheet XLSX: the curl response is 'sheet_selection_required' with a file_key and sheets list. Pass file_key and sheet to question_to_map (or any other tool that accepts file_key) — do not call upload_file again.

After upload, pass the returned file_key to question_to_map to render a map from the full dataset. The file_key is reusable — for additional questions on the same file, pass the same file_key again without re-uploading. When the upload response includes 'guidance', use its next_actions directly for a zero-round-trip render.

Input parameters:

- `file_key` (string)
- `file_name` (string, required)
- `file_size` (integer, required)
- `question` (string)
- `sheet` (string)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/ai-latlong-map-mcp/geocp#diagnostics

## Score history

- 2026-10-02: 57
- 2026-10-01: 56
- 2026-09-30: 56
- 2026-09-29: 55
- 2026-09-28: 55
- 2026-09-27: 55
- 2026-09-26: 55
- 2026-09-25: 55
- 2026-09-24: 45
- 2026-09-23: 54
- 2026-09-22: 54
- 2026-09-21: 54
- 2026-09-20: 54
- 2026-09-19: 54
- 2026-09-18: 54
- 2026-09-17: 54
- 2026-09-16: 54
- 2026-09-15: 54
- 2026-09-14: 54
- 2026-09-13: 54
- 2026-09-12: 54
- 2026-09-11: 54
- 2026-09-10: 54
- 2026-09-09: 54
- 2026-09-08: 54
- 2026-09-07: 54
- 2026-09-06: 54
- 2026-09-05: 54
- 2026-09-04: 56
- 2026-09-03: 55

## Common questions

### What is the LatLong Maps MCP server?

LatLong Maps is an MCP server listed in the public MCP registry as ai.latlong/map-mcp. Turn CSV or Excel data into styled choropleth and categorical maps of India from plain language. This page covers its hosted endpoint (https://geocp.latlong.ai/mcp/).

### Is the LatLong Maps MCP server safe to use?

LatLong Maps scores 57 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the LatLong Maps MCP server expose?

LatLong Maps exposes 5 tools: geocode, get_capabilities, question_to_map, reverse_geocode, upload_file. Their descriptions and schemas cost roughly 1,504 tokens of context every time the server is loaded.

### Does the LatLong Maps MCP server require authentication?

No. We connected to LatLong Maps without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the LatLong Maps MCP server still maintained?

LatLong Maps is still listed as active in the MCP registry. We last reached this channel on 2 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://geocp.latlong.ai/mcp/
- Authorisation metadata: https://geocp.latlong.ai/.well-known/oauth-protected-resource/mcp
- Website: https://latlong.ai/
- Changelog RSS feed: https://verifymcp.io/servers/ai-latlong-map-mcp/geocp.xml
- Changelog JSON feed: https://verifymcp.io/servers/ai-latlong-map-mcp/geocp.json
- HTML version of this page: https://verifymcp.io/servers/ai-latlong-map-mcp/geocp