Provides an AI client with live NFL fantasy football analytics, including player stats, dynasty values, Sleeper league data, and custom Excel projections.
Also known as dachhack/stathead
- Stars
- 3
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- 1.0.126
- Last commit
- Nov 15, 2026
- License
- MIT
What is StatHead MCP?
StatHead provides an AI client with live NFL fantasy football analytics, including player stats, dynasty values, Sleeper league data, and custom Excel projections.
Summary generated from the project's README
About StatHead MCP server
Fantasy football enthusiasts and analysts can access live NFL data and dynasty values directly through AI clients like Claude Desktop, Cursor, or Claude Code. This server provides over 50 tools that cover everything from player season stats and snap counts to advanced Next Gen Stats and injury reports. It connects deeply with Sleeper leagues, allowing users to inspect standings, rosters, transactions, and even cross-league manager exposure.
Users can manage their own custom rankings and projections by exporting data to Excel, making manual edits, and re-importing the files to override the server's internal models. The setup is straightforward using npx for local execution via stdio, or it can be used as a remote connector through a hosted URL. While most data comes from open upstreams like nflverse and Sleeper, specific features like betting odds require an external API key.
What you can do with it
- 1Analyze Sleeper league activityAsk the AI to review recent trades and waiver wire moves within a specific Sleeper league.
- 2Evaluate dynasty trade valueDetermine the fair market value of a player using historical Sleeper trade data and future forecasts.
- 3Assess weekly matchup difficultyIdentify which players have the easiest upcoming schedules based on defensive points allowed.
Written from the project's README. Every use case is checked against a quote from it.
How to install StatHead MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall StatHead MCP in Claude Code
$ claude mcp add stathead-mcp -- npx -y stathead-mcpInstall StatHead MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"stathead-mcp": {
"command": "npx",
"args": [
"-y",
"stathead-mcp"
]
}
}
}Install StatHead MCP in VS Code
.vscode/mcp.json
{
"servers": {
"stathead-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"stathead-mcp"
]
}
}
}Install StatHead MCP in VS Code CLI
$ code --add-mcp '{"name":"stathead-mcp","type":"stdio","command":"npx","args":["-y","stathead-mcp"]}'Install StatHead MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"stathead-mcp": {
"command": "npx",
"args": [
"-y",
"stathead-mcp"
]
}
}
}Install StatHead MCP in Codex
$ codex mcp add stathead-mcp -- npx -y stathead-mcpInstall StatHead MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.stathead-mcp] command = "npx" args = ["-y", "stathead-mcp"]
Install StatHead MCP in Gemini CLI
$ gemini mcp add -s user stathead-mcp npx -- -y stathead-mcpInstall StatHead MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"stathead-mcp": {
"command": "npx",
"args": [
"-y",
"stathead-mcp"
]
}
}
}Automated checks
Ran Nov 14, 2026Full report →Passed our automated checks. That is not a security audit: read the code and give the server the narrowest access it needs. How checks work
- Source repository linked and reachabledachhack/stathead is public and reachablegithub.com/dachhack/stathead ↗Pass
- Package names this repository as its sourcestathead-mcp names dachhack/stathead as sourcenpmjs.com/package/stathead-mcp ↗Pass
- Build provenance (Sigstore attestation)Sigstore attestation for stathead-mcpnpmjs.com/package/stathead-mcp/v/1.0.126#provenance ↗Pass
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-11-12github.com/dachhack/stathead ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNo popular package one edit away from stathead-mcpPass
- No critical advisory in the package or its dependencies1 resolved dependencies checked; no critical advisoryosv.dev ↗Pass
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubstathead-mcp
An MCP server that exposes StatHead's NFL fantasy-analytics toolset to Claude (Desktop / Code), Cursor, or any MCP client — so you can ask natural-language questions and get live, structured NFL data back.
It ships as a single self-contained bundle with no runtime dependencies and
fetches all data on demand, so there's nothing to download or build — npx and
go.
What you get
50 tools spanning player and team data, including:
| Area | Tools |
|---|---|
| Discovery | get_metadata — sources, season coverage, and valid enums (call this first to scope a question) |
| Players | season stats, weekly/game logs, snap counts, injuries, advanced stats, Next Gen Stats, rosters, contracts, get_depth_charts (slot-aware team depth charts with roster status, StatHead depth order and view=changes for moves) |
| Games & plays | schedules/results, play-by-play, participation, FTN charting |
| Fantasy | StatHead blended dynasty values, get_trade_values (StatHead values fitted to millions of completed Sleeper dynasty trades: as-of weekly values with 90% ranges, movers, liquidity, scoring re-pricing, 1/3/6/12-month forecasts, rookie-pick classes and history), get_projections (StatHead's in-house season PPG model), get_weekly_projections (per-week matchup-adjusted points, weeks 1–18), get_matchups (weekly strength of matchup: fantasy points each defense allows per game to QB/RB/WR/TE, ranked 1–32 in your scoring and over expected, by week, per player, rest of schedule, or split by metric such as TE receptions or QB rushing), ADP (FFC / ESPN / Sleeper), get_adp_with_results (ADP vs actual finish), trending adds/drops |
| Sleeper leagues | get_sleeper_user_leagues, get_sleeper_league (standings + rosters), get_sleeper_league_users (cheap manager list), get_sleeper_matchups, get_sleeper_transactions (trades/waivers/adds), get_sleeper_waiver_wire (free agents × trending × projections), get_sleeper_draft, get_sleeper_user_snooper (cross-league exposure), get_sleeper_user_history (multi-season records + titles) — a connected graph: a username lists leagues, a league name (or id) opens that league, and each manager's owner_id/display name walks back to their leagues, exposure, and history |
| Draft & college | get_prospect_outcomes (calibrated boom/bust probabilities + grades), draft picks, get_rookie_class (draft+combine+rookie stats in one call), prospect profiles, combine, college stats (player or cohort), QBR |
| Model & methodology | get_model_docs (how the models work + per-position feature importance), get_player_features (why a player is scored the way they are — feature drivers, bands, direction) |
| Your own numbers | export_excel / import_excel / clear_overrides — download a styled projections / rankings / rookie-rankings workbook (or a by_team projection sheet with live PPR formulas), edit it, re-upload it, and have your values drive every later query |
Bring your own projections & rankings
The same boards the website lets you download are available over MCP as real
.xlsx files (built with zero extra dependencies — Node's own zlib):
export_excel kind="projections"(orrankings,rookie_rankings) writes a styled workbook to disk and returns its path. Edit the highlighted column — Proj PPG for projections, My Rank for rankings/rookies — in Excel or Google Sheets.import_excel path="…"reads it back and saves your values as overrides.get_projections,get_fantasy_rankings, andget_prospect_outcomesthen automatically use your numbers (flagged in the output) until you runclear_overrides.
Workbooks default to $STATHEAD_DIR (or the current working directory); pass
path to choose where they land. Overrides persist in
$STATHEAD_DIR/~/.stathead/overrides.json.
Quick start
Run it directly — no install, no clone:
npx -y stathead-mcp
It speaks MCP over stdio, so you normally wire it into a client rather than run it by hand.
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json
(macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"stathead": {
"command": "npx",
"args": ["-y", "stathead-mcp"]
}
}
}
Claude Code
claude mcp add stathead -- npx -y stathead-mcp
Cursor / other MCP clients
Use the same command/args (npx -y stathead-mcp) in the client's MCP
config.
Claude mobile apps / claude.ai web / Claude Desktop (remote connector)
Prefer a URL over a local command? The same tools are hosted over MCP's
Streamable HTTP transport by a Cloudflare Worker (workers/stathead-mcp/,
deployed via deploy-workers.yml):
https://stathead-mcp.dachhack.workers.dev
Add it under Settings → Connectors → Add custom connector, paste the URL,
and save — no command, no API key. (A GET to that URL returns a small health
payload; clients POST JSON-RPC to it.) The npm/stdio package above stays the
path for Claude Code and Cursor.
The Worker records a privacy-respecting usage counter (MCP method + tool name
only — no IPs or request bodies) to a Cloudflare Analytics Engine dataset
(stathead_mcp_usage). View aggregates at GET /stats once you set the
CF_ACCOUNT_ID + CF_ANALYTICS_TOKEN worker secrets (wrangler secret put …),
or query the dataset in the Cloudflare dashboard. npm download counts don't see
remote-connector usage — this does.
Data sources & freshness
Most tools fetch live from open upstreams (nflverse, Sleeper, FantasyFootballCalculator, ESPN's public endpoints). A handful read committed daily snapshots (StatHead's blended dynasty values, ADP history, in-house season projections), which the server fetches from the hosted StatHead site by default.
Please review DATA_SOURCES.md before redistributing
any data this server returns — several upstreams are proprietary or paid and
carry their own terms.
Configuration
All optional, via environment variables:
| Var | Default | Purpose |
|---|---|---|
STATHEAD_DATA_BASE |
GitHub raw (data branch) | Base URL for committed snapshot data (<base>data/<file>). Defaults to GitHub raw, which serves every file uncompressed/reliably (the Cloudflare host gzips files over its 25 MiB cap, which Node can't decode). Override to point at your own deployment or a local http://localhost:5173/ dev server. |
VITE_KTC_PROXY |
project worker | CORS proxy for KeepTradeCut. Deploy your own (workers/ktc-proxy) to avoid shared rate limits. |
VITE_FC_PROXY |
project worker | CORS proxy for FantasyCalc (workers/fc-proxy). |
VITE_ODDS_API_KEY |
— | Enables betting-odds tools (the-odds-api.com). |
Running from a full repo checkout instead? The server reads snapshots straight
from local public/data/ and ignores STATHEAD_DATA_BASE.
Building from source
The bundle is produced from the StatHead monorepo:
# from the repo root
npm install
npm run build:mcp # → mcp/dist/server.mjs (esbuild, single file)
node mcp/dist/server.mjs
Publishing
Bump version in mcp/package.json and SERVER_VERSION in
mcp/dist/server.mjs (they must match), merge to the default branch, then
any one of:
git push -f origin <commit>:publish-mcp-now # branch-push trigger (works from Claude Code sessions)
git tag mcp-v1.2.3 && git push origin mcp-v1.2.3 # tag trigger
# or Actions → "Publish stathead-mcp to npm" → Run workflow
publish-mcp.yml refuses a version already on npm and a commit not yet on
the default branch, smoke-tests the bundle, publishes via npm Trusted
Publishing, then chains the MCP Registry publish. Manual fallback:
npm run build:mcp # repo root — refresh the bundle
cd mcp && npm publish --access public
The published tarball is just dist/server.mjs + this README (~220 KB).
Listing in the MCP Registry
server.json is the manifest for the official MCP Registry
(io.github.dachhack/stathead-mcp). After an npm publish, the
Publish to MCP Registry workflow (.github/workflows/publish-mcp-registry.yml)
pushes it — authenticating via GitHub OIDC for the io.github.dachhack/*
namespace (no secret). The npm package carries a matching mcpName field so the
registry can verify the package belongs to this entry, and the workflow syncs
server.json's version from package.json so they never drift.
Most community directories (Glama, mcp.so, PulseMCP) mirror the official
registry and/or auto-index public npm packages with the model-context-protocol
keyword (which package.json has), so a registry publish propagates to them.
License
MIT — see ../LICENSE. The code is MIT; the data it returns
is not necessarily redistributable. See
DATA_SOURCES.md.
StatHead MCP questions
How do I add StatHead MCP to Claude Code?
Run claude mcp add stathead-mcp -- npx -y stathead-mcp in your terminal. Add --scope user to make it available in every project.
How do I add StatHead MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
How do I use my own projections and rankings?
You can export projections or rankings to an Excel file, edit the values in Excel or Google Sheets, and then import the file back to use those values as overrides in your queries.
Does this server require any runtime dependencies to run?
No, it is a single self-contained bundle with no runtime dependencies that fetches all data on demand.
Can I use this with Claude Desktop?
Yes, you can add it to your Claude Desktop configuration file using npx to run the server.
How can I access the tools via a URL instead of a local command?
The tools are hosted over MCP's Streamable HTTP transport by a Cloudflare Worker, which you can add as a custom connector in Claude.