Query the FBI's Crime Data Explorer API for US crime statistics, arrest data, hate crimes, NIBRS incidents, and law enforcement data — directly from any MCP-compatible client.
Summarized (SRS) crime data — rates, actuals, and clearances for violent/property crime and subtypes at national, state, or agency level
NIBRS incident-based data for 70+ offense types, plus national estimates by state, region, agency type, or population size
Arrest statistics by offense with optional demographic breakdowns (male, female, race, sex)
Hate crime incidents filterable by 30+ bias motivations
Expanded homicide (SHR) data — victim/offender demographics, weapons, circumstances
Expanded property crime data — stolen/recovered values for burglary, larceny, motor vehicle theft, robbery
Law enforcement data — police employment counts and rates, LEOKA (officers killed/assaulted), LESDC suicides, and use of force incidents
National crime trends across 10 crime types, plus a CDE homepage summary with mission, navigation, and data freshness
Lookup & reference tools — find agencies by state, ORI, or judicial district; look up state lists, offense/bias codes, and refresh dates
Operational utilities — automatic yearly aggregation (with partial-year flags) or monthly granularity, tiered disk caching with manage_cache, spillover file reading for oversized responses, input validation, and built-in rate limiting
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@FBI Crime Data MCP Server what are the crime rates in Texas for 2023?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP (Model Context Protocol) server that provides access to the FBI's Crime Data Explorer API.
Query crime statistics, arrest data, hate crimes, NIBRS incidents, law enforcement employment, and more — directly from any MCP-compatible client.
Created for data journalists, researchers, and anyone interested in exploring US crime data with the power of language models. Initially created for use by the Policy Analyst Agent of the qsv Claude Cowork plugin.
17 tools covering a wide range of crime data topics
Crime trends and Summary Reporting System (SRS) crime data
National Incident Based Reporting System (NIBRS) incident-based data and national estimates
Arrest statistics with demographic breakdowns
Hate crime incidents by bias motivation
Expanded homicide and property crime details
Police employment, Law Enforcement Officers Killed and Assaulted (LEOKA), Law Enforcement Suicide Data Collection (LESDC), and use of force
Agency lookup, reference data, cache management, and spillover reading
Geographic query levels — national, state, and agency for most tools; some also support region (get_police_employment, get_nibrs_estimation) or agency-type / population-size breakdowns (get_nibrs_estimation) — all with automatic parameter validation
Smart yearly aggregation — monthly API data is automatically rolled up into yearly totals (sums for counts, unweighted averages for rates, last value for population), with an option for monthly granularity. Years with fewer than 12 months of data are flagged in a _partial_years key so partial-year sums are never mistaken for annual totals
Tiered disk-backed caching — 90-day time-to-live (TTL) for stable data (trends, reference, summaries, NIBRS estimation), 30-day TTL for dynamic data (incidents, arrests, agency lookups), and 1-day TTL for the homepage summary (refresh dates change frequently). Transient errors (timeouts, HTTP 5xx, rate-limit responses) are never cached
Spillover handling — responses exceeding 128K characters are saved to disk with a preview returned, so large queries are never silently truncated
Input validation — date format/ordering checks, offense and bias code validation, and level-based parameter requirements with clear error messages
Sliding-window rate limiting — 1,000 requests/hour with transparent wait-time feedback
Reference tools for agency lookups (by state, Originating Agency Identifier (ORI), or district with name filtering) and code translations
Related MCP server: mcp-fbi-crime
Get a free API key from api.data.gov
Run with Claude Desktop — add to your claude_desktop_config.json:
{
"mcpServers": {
"fbi-crime-data": {
"command": "uvx",
"args": ["fbi-crime-data-mcp"],
"env": {
"FBI_API_KEY": "your-api-key-here"
}
}
}
}Or run directly:
FBI_API_KEY=your-key uvx fbi-crime-data-mcpTool | Description |
SRS crime data — rates, actuals, clearances for violent crime, property crime, homicide, rape, robbery, assault, burglary, larceny, motor vehicle theft, arson | |
NIBRS incident-based data for 70+ offense types | |
Arrest statistics by offense with optional demographic breakdowns (male, female, sex, race) | |
National crime trend percent changes across 10 crime types | |
NIBRS national estimates by state, region, agency type, or population size |
Tool | Description |
Hate crime incidents by bias motivation (30+ categories) | |
Supplementary Homicide Reports — victim/offender demographics, weapons, circumstances | |
Expanded property crime details — stolen/recovered values for burglary, larceny, motor vehicle theft (MVT), robbery |
Tool | Description |
Officer and civilian employee counts by gender, rates per 1,000 population | |
Officers killed and assaulted — weapons, circumstances, demographics | |
Law enforcement suicide data — demographics, race, duty status, and more | |
Use of force incidents resulting in death, serious injury, or firearm discharge |
Tool | Description |
CDE homepage summary — mission statement, navigation, data freshness, date ranges, and national crime trends |
Tool | Description |
Find law enforcement agencies by state, ORI code, or judicial district | |
State lists, offense/bias code lookups, data refresh dates | |
View cache stats, clear all entries, or clear only expired entries | |
Read spillover files saved when tool responses exceed the size limit |
When a tool response exceeds 128,000 characters, the full result is saved to ~/.cache/fbi-crime-data-mcp/spillover/ and a truncated preview is returned with the file path. To avoid this, narrow your query (shorter date range, specific state/agency).
The preview, not the full payload, is what gets cached, so the cache and the spillover directory belong together. Clear both with manage_cache action="clear" rather than deleting the spillover directory by hand; otherwise cached previews point at files that no longer exist until the cache entry expires.
All data comes from the FBI's Crime Data Explorer API, which provides Uniform Crime Reporting (UCR) data including both the Summary Reporting System (SRS) and the National Incident-Based Reporting System (NIBRS).
Registered key: 1,000 requests per hour (rolling window)
DEMO_KEY: 30 requests per IP per hour
The server includes a built-in rate limiter (1,000 req/hr). The DEMO_KEY limit is enforced API-side.
For more info on the FBI's Crime Data API, see https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/docApi.
# Install dependencies
uv sync
# Run the server locally
FBI_API_KEY=your-key uv run fbi-crime-data-mcp
# Run tests (offline by default — never hits the network)
uv run pytest
# Run integration tests against the live API.
# DEMO_KEY works, but is capped at 30 requests per IP per hour.
FBI_API_KEY=your-key uv run pytest -m integrationMIT
Get arrest statistics by offense, optionally broken down by demographics.
Args: offense: Arrest offense code (e.g., "all", "11" for murder, "30" for robbery, "150" for drug abuse). Use get_reference_data with data_type="offenses" and offense_type="arrest" for full list. level: Geographic level — "national", "state", or "agency" data_type: "counts" for time series or "totals" for demographic breakdowns from_date: Start date in mm-yyyy format (e.g., "01-2020") to_date: End date in mm-yyyy format (e.g., "12-2022") state: Two-letter state abbreviation (required when level is "state") ori: Agency ORI code (required when level is "agency") category: Optional demographic breakdown — "male", "female", "race", or "sex" aggregate: Aggregation level — "yearly" (default, sums monthly into yearly) or "monthly" (monthly granularity). Only applies when data_type is "counts". Years with fewer than 12 months of data are flagged in a top-level "_partial_years" key (rates are unweighted averages of the available months).
Inputs: level (required), offense (required), to_date (required), data_type (required), +5 more. Returns: result
Get a summary of the FBI Crime Data Explorer (CDE) homepage.
Returns the CDE mission statement, navigation structure, data freshness (last refresh dates), available data date ranges, and the full national crime trends response (percent changes across 10 crime types, matching the trends section displayed on the CDE homepage). Provides orientation on what the CDE offers and how current its data is.
Inputs: no named parameters. Returns: result
Get national crime trend data showing percent changes across 10 crime types (murder, rape, robbery, aggravated assault, violent crime, burglary, larceny, motor vehicle theft, arson, property crime).
Args: from_year: Start year in yyyy format (e.g., "2015"). Optional. to_year: End year in yyyy format (e.g., "2022"). Optional.
Inputs: to_year, from_year. Returns: result
Get Supplementary Homicide Report (SHR) data with expanded details on homicides including victim/offender demographics, weapons, and circumstances.
Args: level: Geographic level — "national", "state", or "agency" data_type: "counts" for time series or "totals" for demographic/detail breakdowns from_date: Start date in mm-yyyy format (e.g., "01-2020") to_date: End date in mm-yyyy format (e.g., "12-2022") state: Two-letter state abbreviation (required when level is "state") ori: Agency ORI code (required when level is "agency") aggregate: Aggregation level — "yearly" (default, sums monthly into yearly) or "monthly" (monthly granularity). Only applies when data_type is "counts". Years with fewer than 12 months of data are flagged in a top-level "_partial_years" key (rates are unweighted averages of the available months).
Inputs: level (required), to_date (required), data_type (required), from_date (required), +3 more. Returns: result
Get expanded property crime data with details on stolen/recovered property values.
Args: offense: Property offense code — "NB" (Burglary), "NL" (Larceny), "NMVT" (Motor Vehicle Theft), "NROB" (Robbery). level: Geographic level — "national", "state", or "agency" data_type: "counts" for time series or "totals" for property value breakdowns from_date: Start date in mm-yyyy format (e.g., "01-2020") to_date: End date in mm-yyyy format (e.g., "12-2022") state: Two-letter state abbreviation (required when level is "state") ori: Agency ORI code (required when level is "agency") aggregate: Aggregation level — "yearly" (default, sums monthly into yearly) or "monthly" (monthly granularity). Only applies when data_type is "counts". Years with fewer than 12 months of data are flagged in a top-level "_partial_years" key (rates are unweighted averages of the available months).
Inputs: level (required), offense (required), to_date (required), data_type (required), +4 more. Returns: result
Get hate crime statistics, optionally filtered by bias motivation. Returns incident counts, victim types, offense types, offender demographics, and locations.
Args: level: Geographic level — "national", "state", or "agency" from_date: Start date in mm-yyyy format (e.g., "01-2020") to_date: End date in mm-yyyy format (e.g., "12-2022") bias: Bias code to filter by (e.g., "12" for Anti-Black, "21" for Anti-Jewish, "24" for Anti-Islamic). Use get_reference_data with offense_type="hate-crime" for full list. If omitted, returns all biases. data_type: "counts" for time series or "totals" for aggregate data (default: "counts") state: Two-letter state abbreviation (required when level is "state") ori: Agency ORI code (required when level is "agency") aggregate: Aggregation level — "yearly" (default, sums monthly into yearly) or "monthly" (monthly granularity). Only applies when data_type is "counts". Years with fewer than 12 months of data are flagged in a top-level "_partial_years" key (rates are unweighted averages of the available months).
Inputs: level (required), to_date (required), from_date (required), ori, +4 more. Returns: result
Get LEOKA data on law enforcement officers killed and assaulted, including weapons used, circumstances, officer demographics, and offender demographics.
Args: report_type: "monthly" for a specific month or "ytd" for year-to-date summary year: Year for the data (e.g., 2022) month: Month number 1-12 (January=1, December=12). Required when report_type is "monthly".
Inputs: year (required), report_type (required), month. Returns: result
Get law enforcement suicide data with breakdowns by demographics, race, location, duty status, and more.
Args: year: Year for the data (e.g., 2022) chart_type: Chart/breakdown type. Use get_reference_data or see error message for valid values.
Inputs: year (required), chart_type (required). Returns: result
Get NIBRS incident-based crime data for 70+ offense types.
Args: offense: NIBRS offense code (e.g., "13A" for aggravated assault, "09A" for murder, "11A" for rape, "120" for robbery, "220" for burglary). Use get_reference_data for full list. level: Geographic level — "national", "state", or "agency" from_date: Start date in mm-yyyy format (e.g., "01-2020") to_date: End date in mm-yyyy format (e.g., "12-2022") data_type: "counts" for time series data or "totals" for aggregate breakdowns (default: "counts") state: Two-letter state abbreviation (required when level is "state") ori: Agency ORI code (required when level is "agency") aggregate: Aggregation level — "yearly" (default, sums monthly into yearly) or "monthly" (monthly granularity). Only applies when data_type is "counts". Years with fewer than 12 months of data are flagged in a top-level "_partial_years" key (rates are unweighted averages of the available months).
Inputs: level (required), offense (required), to_date (required), from_date (required), +4 more. Returns: result
Get NIBRS national crime estimates derived from incident data. Supports breakdowns by state, region, agency type, and population size.
Args: offense: NIBRS offense code (e.g., "13A", "09A"). Use get_nibrs_data docs for codes. level: "national", "state", "region", "agency-type", or "size" year: Year for estimation data state: Two-letter state abbreviation (required when level is "state") region: Region code — "M" (Midwest), "N" (Northeast), "S" (South), "W" (West). Required when level is "region". agency_type: "S" (Size) or "T" (Agency Type). Required when level is "agency-type" or "size". agency_type_location: "C" (City) or "N" (County). Required when level is "agency-type". size_group: Size group "1"-"8" (1=Cities 250K+, 6=Cities under 10K, 7=MSA Counties, 8=Non-MSA Counties). Required when level is "size".
Inputs: year (required), level (required), offense (required), state, +4 more. Returns: result
Get law enforcement employee data including officer/civilian counts by gender, rates per 1,000 population.
Args: level: Geographic level — "national", "state", "agency", or "region" from_year: Start year in yyyy format (e.g., "2015") to_year: End year in yyyy format (e.g., "2022") state: Two-letter state abbreviation (required for "state" and "agency" levels) ori: Agency ORI code (required for "agency" level) region: Region name — "midwest", "south", "northeast", or "west" (required for "region" level)
Inputs: level (required), to_year (required), from_year (required), ori, +2 more. Returns: result
Get reference data: state lists, offense/bias code lookups, data properties, or data refresh dates.
Args: data_type: Type of reference data — "states" (list of states/territories), "offenses" (offense or bias code lookup), "cde_properties" (data date ranges), or "refresh_date" (last data refresh dates) offense_type: Required when data_type is "offenses" — "arrest" for arrest offense codes or "hate-crime" for hate crime bias codes
Inputs: data_type (required), offense_type. Returns: result
Get summarized (SRS) crime data including offense rates, actuals, clearances, and population coverage.
Args: offense: SRS offense code (e.g., "V" for Violent Crime, "P" for Property Crime, "HOM", "RPE", "ROB", "ASS", "BUR", "LAR", "MVT", "ARS"). Use get_reference_data for full list. level: Geographic level — "national", "state", or "agency" from_date: Start date in mm-yyyy format (e.g., "01-2020") to_date: End date in mm-yyyy format (e.g., "12-2022") state: Two-letter state abbreviation (required when level is "state") ori: Agency ORI code (required when level is "agency") aggregate: Aggregation level — "yearly" (default, sums monthly into yearly) or "monthly" (monthly granularity). Years with fewer than 12 months of data are flagged in a top-level "_partial_years" key (rates are unweighted averages of the available months).
Inputs: level (required), offense (required), to_date (required), from_date (required), +3 more. Returns: result
Get use of force data from law enforcement agencies. Covers incidents resulting in death, serious bodily injury, or firearm discharge.
Args: report_type: "summary" for participation/submission data, "questions" for detailed question data, "reports" for report data year: Year for the data. Required for "summary" and "questions" types. location: "national" or a two-letter state abbreviation. Required for "summary" type. group: Group identifier for "questions" or "reports" types. quarter: Quarter (1-4) for "questions" type. spec: Report specification for "reports" type.
Inputs: report_type (required), spec, year, group, +2 more. Returns: result
Look up law enforcement agencies by state, ORI code, or judicial district code.
Args: lookup_type: How to look up — "by_state" (list agencies in a state), "by_ori" (specific agency by ORI), or "by_district" (agencies by judicial district code) state: Two-letter state abbreviation (required for by_state and by_ori) ori: Agency ORI identifier (required for by_ori) district_code: Judicial district code (required for by_district) name_filter: Optional substring to filter results by agency name (case-insensitive). Only applies to by_state and by_district lookups. offset: Number of results to skip (for pagination). Applied after name_filter. limit: Maximum number of results to return (for pagination). Applied after name_filter.
Inputs: lookup_type (required), ori, limit, state, +3 more. Returns: result
Manage the FBI Crime Data response cache.
Args: action: "status" (show cache stats), "clear" (wipe entire cache), or "clear_expired" (remove only expired entries)
Inputs: action (required). Returns: result
Read a spillover file that was saved when a tool response exceeded the size limit.
Use this to retrieve data beyond the preview shown in a truncated response. The filename is provided in the spillover notice (e.g., "get_nibrs_data_a1b2c3d4.json").
Args: filename: Name of the spillover file (e.g., "get_nibrs_data_a1b2c3d4.json"). Use "list" to see all available spillover files. offset: Character position to start reading from (default: 0). limit: Maximum number of characters to return (default: 50000, max: 100000).
Inputs: filename (required), limit, offset. Returns: result
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 17 tools
Most tools map to distinct FBI datasets or utilities, but get_nibrs_data vs get_nibrs_estimation and get_summarized_crime_data vs get_crime_trends require careful reading to avoid misselection. Descriptions generally clarify scope.
All names use snake_case verb_noun structure (get_*, lookup_agency, manage_cache, read_spillover), with a predictable get_ prefix for data retrieval tools. No mixed conventions.
17 tools is on the heavy side for a data-access MCP server, falling in the borderline 16-25 range. Each tool maps to a distinct dataset, but the surface is large enough to feel sprawling.
The server covers the major FBI Crime Data Explorer domains: trends, summarized, NIBRS, NIBRS estimates, arrests, hate crime, expanded homicide/property, LEOKA, police employment, suicide, use of force, agency lookup, reference data, and spillover handling. No obvious lifecycle gaps for a read-only data server.
U.S. civic data for AI agents: reps, votes, bills, finance, lobbying, cited gov sources. 47 tools.
Geocoding, routing, isochrones, traffic, weather, and place search for AI agents. 19 MCP tools.
US government data as clean JSON for AI agents: SAM.gov contract opportunities, USAspending awards, Grants.gov grants, House STOCK Act trades, and SEC EDGAR filings (Form 4 insider trades, 8-K events, 13F holdings, 13D/G stakes, XBRL fundamentals, 10-K/10-Q sections). 19 read-only tools. Data is as fresh as each source publishes; congressional trades lag up to 45 days and report dollar ranges (House only). Free tier, no card.
SEC filings — 13F, Form 4, 8-K, 13D, IPO and short-interest data — as 17 agent tools.
