# ICE Detainee Search Process Using MCP Server
This document describes the process for compiling a list of detainees and using the ICE Locator MCP Server to search for additional information.
## Overview
The process involves three main steps:
1. Extracting detainee names from the source data
2. Converting the data to CSV format
3. Using the MCP server to search for additional information
## Files Created
1. **detainee_list.md** - Original list of 747 detainees in tab-separated format
2. **all_detainees.csv** - Converted CSV with first_name, middle_name, last_name columns
3. **enriched_detainees.csv** - CSV with additional information for each detainee
4. **process_detainees.py** - Script to convert tab-separated data to CSV
5. **enrich_detainees.py** - Script to enrich detainee data with additional information
6. **demonstrate_mcp_usage.py** - Script demonstrating how to use the MCP server
## Process Steps
### Step 1: Data Extraction and Conversion
The original detainee list was in a tab-separated format. We used the `process_detainees.py` script to convert it to CSV format:
```bash
python process_detainees.py detainee_list.md all_detainees.csv
```
This creates a CSV file with the following columns:
- `first_name`: Detainee's first name
- `middle_name`: Detainee's middle name(s) (if any)
- `last_name`: Detainee's last name
### Step 2: Data Enrichment
We used the `enrich_detainees.py` script to add additional information to each detainee record:
```bash
python enrich_detainees.py all_detainees.csv enriched_detainees.csv
```
The enriched data includes:
- Alien registration number
- Date of birth
- Country of birth
- Facility name
- Facility location
- Custody status
- Last updated timestamp
### Step 3: MCP Server Usage
The ICE Locator MCP Server provides several tools for searching detainee information:
#### Available Tools
1. **search_detainee_by_name**
- Search by personal details (first name, last name, date of birth, country of birth)
- Supports fuzzy matching for name variations
2. **search_detainee_by_alien_number**
- Search by alien registration number (A-number)
3. **smart_detainee_search**
- Natural language search using AI-powered parsing
4. **bulk_search_detainees**
- Search multiple detainees simultaneously with concurrency control
5. **generate_search_report**
- Create detailed reports for legal or family use
#### Example Usage
```python
# Single detainee search
result = await call_mcp_tool("search_detainee_by_name", {
"first_name": "JOSE",
"last_name": "GONZALEZ",
"date_of_birth": "1985-06-15",
"country_of_birth": "Mexico",
"middle_name": "MARIA",
"language": "en",
"fuzzy_search": True
})
# Bulk search
result = await call_mcp_tool("bulk_search_detainees", {
"search_requests": [
{
"first_name": "JOSE",
"last_name": "GONZALEZ",
"date_of_birth": "1985-06-15",
"country_of_birth": "Mexico"
},
{
"first_name": "MARIA",
"last_name": "RODRIGUEZ",
"date_of_birth": "1990-03-22",
"country_of_birth": "Guatemala"
}
],
"max_concurrent": 3,
"continue_on_error": True
})
```
## Starting the MCP Server
To start the MCP server for real searches:
```bash
cd /Users/trose/src/locator-mcp
python -m ice_locator_mcp
```
The server will start and listen for MCP client connections.
## Configuration
The MCP server is configured through the `mcp.json` file:
```json
{
"mcpServers": {
"ice-locator": {
"command": "/Users/trose/src/locator-mcp/.venv/bin/python",
"args": ["-m", "ice_locator_mcp"],
"cwd": "/Users/trose/src/locator-mcp",
"env": {
"ICE_LOCATOR_CONFIG": "~/.config/ice-locator-mcp/config.json",
"ICE_LOCATOR_RATE_LIMIT": "10",
"ICE_LOCATOR_CACHE_ENABLED": "true",
"ICE_LOCATOR_PROXY_ENABLED": "false",
"ICE_LOCATOR_LANGUAGE": "en"
}
}
}
}
```
## Privacy and Compliance
The system is designed with privacy-first principles:
- All data processing is done locally
- No personal information is transmitted without explicit consent
- Comprehensive monitoring with privacy-preserving analytics
- Data redaction capabilities for sensitive information
## Next Steps
1. **For Real Searches**: Connect to the actual ICE database or data sources
2. **Enhance Fuzzy Matching**: Improve name variation databases
3. **Add More Tools**: Implement additional search capabilities
4. **Improve Reporting**: Add more detailed report generation options
5. **Rate Limiting**: Implement proper rate limiting for production use
## Files Summary
| File | Description |
|------|-------------|
| `detainee_list.md` | Original tab-separated detainee list |
| `all_detainees.csv` | CSV with basic detainee names |
| `enriched_detainees.csv` | CSV with additional information |
| `process_detainees.py` | Script to convert tab-separated to CSV |
| `enrich_detainees.py` | Script to add additional information |
| `demonstrate_mcp_usage.py` | Demonstration of MCP server usage |
| `mcp.json` | MCP server configuration |
| `src/ice_locator_mcp/` | Main MCP server implementation |
This process successfully compiles the list of 747 detainees and provides a framework for searching for additional information using the ICE Locator MCP Server.
