Ice Locator

Provides an AI-powered MCP server for searching ICE detainee locations with multilingual natural language queries and heatmap visualization.

python

5

GitHub Stars

python

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "trose-ice-locator-mcp": {
      "command": "ice-locator-mcp",
      "args": []
    }
  }
}

You set up and run an MCP server that lets you search ICE detainee locations using natural language queries, with features like multilingual support, bulk searches, and legible reports. This server runs locally via a simple command after installation, so you can integrate it with compatible MCP clients to empower your investigations, legal work, or advocacy efforts.

How to use

To use the ice_locator MCP server, install it locally, then run the provided runtime command. Connect an MCP client to this server to issue natural language queries, start bulk searches, and generate legal reports. You can perform searches by name or by specific identifiers, or use AI-powered natural language queries to analyze multiple detainees at once. The web app delivers heatmap visualizations of facility locations for quick insight.

How to install

Prerequisites: you need Python and pip installed on your system.

pip install ice-locator-mcp

After installation, start the MCP server using the runtime command shown in the configuration snippet.

icelocator
ice-locator-mcp

Additional setup and usage notes

The server exposes a local command you run to start the MCP service. A recommended client invokes the following tools for common tasks: natural language search, name-based search, bulk search, and legal report generation. You can access the interactive web heatmap at the web address provided for visualization.

Example usage patterns include:

# Natural language search
result = await smart_detainee_search(
  query="Find Maria Rodriguez from Guatemala born around 1985"
)

# Name-based search
result = await search_detainee_by_name(
  first_name="John",
  last_name="Doe",
  date_of_birth="1990-01-15",
  country_of_birth="Mexico"
)

# Bulk search
result = await bulk_search_detainees(
  search_requests=[...],
  max_concurrent=3
)

Web app and visualization

View the interactive heatmap and detainee location insights at the web app URL ice-locator-mcp.vercel.app.

Available tools

search_detainee_by_name

Search by personal information using fuzzy matching to locate detainee records by name and related attributes.

search_detainee_by_alien_number

Search by alien registration number (A-number) to retrieve a detainee’s record.

smart_detainee_search

AI-powered natural language search that interprets queries and finds relevant detainee records.

bulk_search_detainees

Execute multiple detainee searches in parallel to improve throughput.

generate_search_report

Create comprehensive legal reports based on search criteria and results.