Fill My Bracket
com.fillmybracket/fill-my-bracketfillmybracket.comstreamable-httpv1.2.1Scanned
March Madness bracket help: picks, matchup analysis, and upset guidance for your pool.
4 tools: 4 only read.
- Reads private data · 0
- Reads text others write · 0
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
Anlyon alternative
Run the operations that matter as governed actions.
Declare what an action changes, and Anlyon puts it under a shared limit, binds every approval to the exact request, and grades the outcome. It works on any HTTPS API you can describe as an action.
Anlyon does not run, proxy or govern this MCP server. A tool your own code calls directly is not routed through Anlyon and is not governed by it.
Import an OpenAPI spec →Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
4 tools
Only reads 4Show 4 read-only tools
get_backtestFor a past men's tournament (2016-2025; no tournament was played in 2020), returns the winning recipes: the site's own five-question engine was run over that year's REAL field (rea…No findings
Why it is in this group
- The name "get_backtest" says "get": a word for reading.
Description, as the server wrote it For a past men's tournament (2016-2025; no tournament was played in 2020), returns the winning recipes: the site's own five-question engine was run over that year's REAL field (real teams, real seeds) across 108 answer combinations, and every bracket was scored against what actually happened under standard pool scoring (1-2-4-8-16-32, max 192). With no style inputs, returns the year's real champion and Final Four, the best total score and the recipe that earned it, the best first-weekend score, how many recipes ended with the real champion, and how many reproduced the full Final Four (in most years, none did). With style inputs (chaos, tiebreak, champion_pick, loyalty), scores that exact recipe instead. Counted history, not a prediction: a recipe is what WOULD have scored in a finished tournament, and the result says so. No odds, no betting advice.
Annotations the server declares
- title: Get backtested winning recipes
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_cinderella_radarReturns Fill My Bracket's Cinderella radar, as of Sunday, November 1, 2026, exactly as published on the site's rankings page: the unranked teams the model rates far above their pol…No findings
Why it is in this group
- The name "get_cinderella_radar" says "get": a word for reading.
Description, as the server wrote it Returns Fill My Bracket's Cinderella radar, as of Sunday, November 1, 2026, exactly as published on the site's rankings page: the unranked teams the model rates far above their poll votes (each with its league, AP votes, model rank, and a plain why line), the Cinderella watch list, and the ranked teams flagged as vulnerable favorites on their projected seed lines, each with that line's counted first-round record from 1985-2025. These are November signals from the preseason poll and one model, stated as the site states them: not predictions, not odds, and no guarantee any of these teams makes the 2027 field, which is set on Selection Sunday, March 14, 2027.
Annotations the server declares
- title: Get the Cinderella radar
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_rankingsReturns the current college basketball rankings exactly as published on fillmybracket.com/college-basketball-rankings: the Preseason AP Top 25 (released Monday, October 19, 2026, w…No findings
Why it is in this group
- The name "get_rankings" says "get": a word for reading.
Description, as the server wrote it Returns the current college basketball rankings exactly as published on fillmybracket.com/college-basketball-rankings: the Preseason AP Top 25 (released Monday, October 19, 2026, with first-place votes), the other teams receiving votes, and the Bart Torvik T-Rank model ranks (preseason baselines, read November 1, 2026). Rankings as of Sunday, November 1, 2026, and every result says so. With no argument, returns the full Top 25 plus the model's own top 10, so the poll-vs-model gaps are visible. With a team name, returns that team's AP rank or votes, its model rank, and any radar or vulnerable-favorite note the site carries for it. A team outside the published data gets a plain not-found: nothing is invented. Real teams and real ranks only; this is current-season data, not the 2027 tournament field, which is set on Selection Sunday, March 14, 2027.
Annotations the server declares
- title: Get current rankings
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_upset_countsAnswers upset questions from the men's tournament counted game by game, 1985-2025 (40 tournaments, 2,520 games): how often a seed line wins its first-round game (stated as wins in …No findings
Why it is in this group
- The name "get_upset_counts" says "get": a word for reading.
Description, as the server wrote it Answers upset questions from the men's tournament counted game by game, 1985-2025 (40 tournaments, 2,520 games): how often a seed line wins its first-round game (stated as wins in 160 games), how many first-round upsets a typical year has, and how far a seed has gone. Ask with a matchup ('12 vs 5'), a seed, or no argument for the whole table. These are counted historical facts, not odds, not win probabilities for any future game, and not betting advice. The dataset is men's only and covers the main bracket from the round of 64 (no First Four games).
Annotations the server declares
- title: Get counted upset history
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or untrusted-input or outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write0
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 5, 2026, 03:18 UTC, unchanged since Nov 5, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 0
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 1.2.1. Values in angle brackets are yours to fill in.
claude mcp add --transport http fill-my-bracket https://mcp.fillmybracket.com/mcpclaude mcp add --transport http fill-my-bracket https://mcp.fillmybracket.com/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__fill-my-bracket__get_rankings",
"mcp__fill-my-bracket__get_cinderella_radar",
"mcp__fill-my-bracket__get_upset_counts",
"mcp__fill-my-bracket__get_backtest"
],
"ask": [],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as fill-my-bracket. Review it before you use it.
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://anlyon.dev/api/v1/servers/com.fillmybracket/fill-my-bracket/verify?hash=889acfb1750dbb4a307e48b89908281576c9eaf16842423276202fde1e905bdc" | jq -e '.changed == false'Pinned hash 889acfb1750d, from Nov 5, 2026, 04:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://anlyon.dev/servers/com.fillmybracket/fill-my-bracket/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://anlyon.dev/servers/com.fillmybracket/fill-my-bracket)Source: MCP Registry · repository. Synced Nov 5, 2026, 04:18 UTC.