Skip to content
FDB4 logo

FDB4 MCP server

Which servers in the official MCP registry are FDB4's own, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-20

← Back to FDB4

Whose server it is
  • Registry nameio.fdb4/football-database-and-resultsverified in the MCP registry

The registry name sits under fdb4.io. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-20
  • Stateanswerschecked 2026-10-20
  • Tools11
  • Server calls itselfFootball Database and Results
  • Server version1.2.0
  • Protocol version2025-06-18

On 2026-10-20 the server answered initialize and tools/list without a sign-in. First answer in 183 ms.

Should an agent connect itNot proven either way yet
  • Whether its tools do the product's job has not been judged yet.
  • Checked on 3 of the 7 days it takes to judge reliability.

Once connected

  • For "Browse results and head-to-head records between two teams", the likely first call is head_to_head.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 3, the first on 2026-10-20
  • INITE MCP scoreafter a weeka score needs 7 daily checks: 3 so far
  • Answeredmeasuring3 of the 7 daily checks it takes to print a share
  • Median first answer537 ms
  • Space in an agent's context≈ 3,523 tokensthe definitions of all its tools (11), which an agent carries on every turn
  • Tool definitionsA · 100 of 100
  • Typed results11 of 11 toolsdeclared outputSchema: results come typed, read once
  • Every tool passes all six checks.

What an agent can do with it

  • not sure · Find teams on the longest active run of non-draw results
  • can do · Browse results and head-to-head records between two teams
  • not sure · Pull match data into scripts through the REST API

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

Worth a look

None of the five warning signs in these definitions.

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (30%), doing the product's job (25%), definitions (20%), context (10%) and use (15%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers11 tools · read 2026-10-20
  • predict_draws

    [PAID] Rank teams by how overdue a draw is — the question this database exists to answer. Every team has a historical draw rate and a run of matches since its last draw. When the run is long relative to how often that team normally draws…

  • get_team_prediction

    [PAID] How overdue a draw one named team is, plus the results that made it so. Returns the team's draw-likelihood rank and score among all ranked teams, the signals behind it (draw rate, current no-draw streak, longest ever, draw debt)…

  • get_draw_model

    [PAID] Every draw reading FDB4 holds for ONE match — use it when someone asks how likely a specific fixture is to end level, rather than which teams are overdue a draw (`predict_draws` answers that). Five readings come back side by side…

  • top_streaks

    [PAID] The all-time leaderboard of no-draw runs — the longest any team has gone without drawing. Answers "what is the longest run of matches without a draw on record?" and "who else has done it?". Each entry is one run: its length, the…

  • get_team_streaks

    [FREE] Every no-draw run one team has ever had, plus the one it is on now. A no-draw run is a sequence of matches that ended in a win or a loss. This returns the run currently in progress with each match in it — date, competition…

  • list_teams

    [FREE] Find a team and its id — the lookup every other team tool depends on. Search by name prefix with `q` ("arse" finds Arsenal), or browse by first letter, or list one kind of team. Each row carries the numeric id and the slug that…

  • get_team

    [FREE] One team's profile and its all-time draw statistics. Returns the team's name, country and type alongside the numbers this database is built on: matches played, draws, draw percentage, the no-draw streak currently running, the…

  • list_tournaments

    [FREE] Every competition covered — leagues, domestic cups, continental cups, World Cups — with how many matches and how many draws each one holds. Two uses. Without `tournament` it answers "what is in this database, and how draw-heavy is…

  • list_countries

    [FREE] Every country in the database, with its FIFA code and id. Use it to turn a country name into the id and FIFA code the rest of the data is keyed by, or to check which countries are covered at all. Search with `q`, which matches a…

  • search_results

    [FREE for the last 7 days · PAID for the archive] Search the match archive: finished matches, newest first, with scores. This is the raw history behind every other tool — decades of results across national team football, domestic leagues…

  • head_to_head

    [PAID] The complete record between two teams, and every match they have ever played. Answers "who wins more between these two, and how often do they draw?" — matches played, wins each way, draws, goals each way and the draw percentage…

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "football-database-and-results": {
      "type": "http",
      "url": "https://fdb4.io/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http fdb4 https://fdb4.io/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://fdb4.io/mcp as a custom connector in the settings.

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.