mcpField
B
Ball Ranks Fantasy Sports

Ball Ranks Fantasy Sports

Remote
ballranks.comGames & Entertainment0Updated 1w ago

About Ball Ranks Fantasy Sports

Ball Ranks is a read-only MCP server for published fantasy football and basketball rankings and player projections. It serves structured results with source links for fantasy sports players and compatible agents.

Key features

  1. 01Get weekly rankings
  2. 02Get season rankings
  3. 03Get player rankings
  4. 04Get player projections
  5. 05Return structured results with source links

Use cases

  1. 01Check weekly fantasy rankings
  2. 02Compare season-long player rankings
  3. 03Review a player's projection
  4. 04Trace a result to its source

MCP support

Exposes four read-only ranking and projection tools through a remote MCP endpoint. Clients discover the live tool list; anonymous access is available with lower request limits, while API keys provide account-level limits.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://ballranks.com/api/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

4
ParameterTypeRequiredDescription
weekintegerYes
limitintegerNo
modelenumNov0 | v1
sportenumNobasketball | football
offsetintegerNo
seasonanyNo
positionenumNoQB | RB | WR | TE | K | DST
scoring_formatenumNostandard | half_ppr | ppr

Details

From the project README

Ball Ranks plugin

Ball Ranks provides published fantasy basketball and football rankings and player projections through a read-only remote MCP server.

This repository packages the same live server for ChatGPT and Codex, Claude, Google Antigravity, Gemini CLI, Grok Build, Muse Code, and other Agent Plugins or MCP-compatible clients. The initial MCP surface contains four tools; clients discover the live tool list so Ball Ranks can add capabilities without changing the server URL.

Server

Install

Clients that support Agent Plugins can install this repository directly. Platform-native manifests are also included:

  • plugin.json and mcp.json: Agent Plugins, ChatGPT, and Codex
  • server.json: Official MCP Registry
  • .codex-plugin/plugin.json: Codex local plugin compatibility
  • .claude-plugin/plugin.json and .mcp.json: Claude Code and Grok Build
  • antigravity-plugin/: Google Antigravity 2.0, CLI, and IDE
  • gemini-extension.json and GEMINI.md: enterprise and paid API-key Gemini CLI installations

No API keys are stored in this repository. Anonymous Model Zero access works with lower request limits. Ball Ranks API keys can be created at https://ballranks.com/account#developer-access for account-level limits; clients must store credentials securely.

Google Antigravity

Google moved unpaid and Google One Gemini CLI users to Antigravity CLI on June 18, 2026. Clone this repository, then install the native package from its dedicated directory:

agy plugin install ./antigravity-plugin

Inside the Antigravity CLI, the equivalent command is:

/plugin install ./antigravity-plugin

The package installs as ball-ranks, connects to the production MCP endpoint through serverUrl, and includes a focused Ball Ranks skill. Run agy plugin list and open /mcp to verify that the plugin and server loaded.

For a workspace-only Antigravity 2.0 or IDE installation, copy the package directory to .agents/plugins/ball-ranks/. For a global installation, copy it to ~/.gemini/config/plugins/ball-ranks/.

Marketplace publication is separate from the package. Google currently asks publishers to complete its Antigravity Marketplace interest form.

Gemini CLI (enterprise and API-key installations)

Install directly from the public repository:

gemini extensions install https://github.com/Ollynov/ball-ranks-plugin

Restart Gemini CLI after installation, then run /extensions list and /mcp to confirm that ball-ranks and its live tools are available. GEMINI.md gives Gemini concise tool-selection and response guidance. This package remains for enterprise and paid API-key Gemini CLI installations.

The public gallery discovers repositories automatically. This repository is public, has the gemini-cli-extension GitHub topic, and keeps gemini-extension.json at the repository root. No separate gallery submission is required.

Current tools

  • get_week_rankings
  • get_season_rankings
  • get_player_rankings
  • get_player_projections

All tools are read-only and return structured content plus a canonical sourceUrl.

Review material

review/review-cases.json contains the same five positive and three negative cases imported from plugin.json. review/demo-script.md is the shared walkthrough plan. review/openai-submission.md records the remaining dashboard work and tool-annotation justifications. Run the cases against production after any tool-contract change.

Build the OpenAI submission ZIP

Run ./scripts/build-openai-submission.sh from a clean, committed checkout. The script validates the package and creates dist/ball-ranks-openai-<version>.zip from the current commit. Rebuilding the same commit produces the same SHA-256 digest.

The ZIP contains only the portable manifest, remote MCP configuration, onboarding skill, and referenced square icons. It intentionally excludes platform compatibility manifests, review notes, repository documentation, and screenshots.

Support

Email [email protected] or visit https://ballranks.com/support. Never include passwords, sign-in cookies, or full API keys in a support request.

License

The packaging files and skill instructions in this repository are available under the MIT License. Ball Ranks data and services remain subject to the Ball Ranks Terms of Service.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.