scorezilla-mcp
scorezilla-mcp
Official Model Context Protocol (MCP) server for Scorezilla — the easiest way to add a leaderboard to your game. Connect this server to your AI coding assistant (Claude Code, Cursor, Continue.dev, …) and ship a working leaderboard without leaving your editor.
What you can ask the AI to do
"Add a leaderboard to my game" → it bootstraps a game + board and pastes ready-to-run TypeScript SDK code into your project
"What did my last test score rank?" → it reads your live leaderboard
"List my games" / "show me the boards on X" → it inspects what you already have
Six tools total — five read-only, one that creates resources (bootstrap_leaderboard).
Related MCP server: Semantic Pen MCP Server
Install + configure
Status — v0.2.0. Published on the
@latestdist-tag.0.2.0adds the integration-axis arguments (identity strategy, OAuth provider, hosting/anti-cheat pattern, server language) tobootstrap_leaderboard+get_sdk_snippet, so the assistant can generate the drop-in widget embed, the secure server-validated (anti-cheat) path, and OAuth identity — not just the default anonymous + client-only snippet. The tool surface (six tools, auth, env vars, CLI flags) is stable within0.2.xper pre-1.0 semver.
1. Get a token
Sign in at dashboard.scorezilla.dev, open MCP tokens, click Create token. Copy the mcp_live_* value once — it's not shown again.
2. Add the server to your AI coding assistant
Claude Code — edit ~/.claude/settings.json:
{
"mcpServers": {
"scorezilla": {
"command": "npx",
"args": ["-y", "@scorezilla/mcp"],
"env": {
"SCOREZILLA_TOKEN": "mcp_live_…"
}
}
}
}🔒 Keep
~/.claude/settings.jsonprivate. The token is stored in plaintext in that file. Make sure it's not committed to git (it's usually in your.gitignore), not synced to a public dotfiles repo, and not backed up to a shared location. On macOS/Linux:chmod 600 ~/.claude/settings.jsonso only your user can read it. If a token leaks, revoke it at dashboard.scorezilla.dev/account/tokens.
Cursor — open Settings → Features → MCP → Add new MCP server, then use the same command + args + env shape.
Anything else MCP-compatible — point your client at npx -y @scorezilla/mcp with SCOREZILLA_TOKEN set in the environment.
3. Ask away
In Claude Code or Cursor: "Add a Scorezilla leaderboard to this game."
Tools
Tool | What it does |
| Lists your games. Use this first to orient. |
| Lists leaderboards under a game. |
| Returns the public key (safe to embed) and the secret-key prefix. The full secret never leaves the dashboard. |
| Returns the top entries on a board. The "is my integration working?" tool. |
| Returns ready-to-paste integration code for a board. Optional axis args tailor it: anonymous/OAuth identity, client-only vs. server-validated anti-cheat, and the server language (TS/Python/Go/C#). |
| Creates a new game + first board in one call, then returns the widget embed + SDK snippet + a plain-English recommendation. Same optional axis args (anti-cheat, OAuth, server language). The 90-second-demo path. |
Flags
scorezilla-mcp [--read-only] [--base-url=<url>] [--version] [--help]--read-only— refuse to registerbootstrap_leaderboard. Use this on shared/CI configs to guarantee the AI can't create resources.--base-url=<url>— override the API origin. Defaults tohttps://api.scorezilla.dev. Useful for self-hosted or staging environments.
Env vars
SCOREZILLA_TOKEN— required. Bearer token issued at dashboard.scorezilla.dev/account/tokens.SCOREZILLA_BASE_URL— same as--base-url, but via env. CLI flag wins if both are set.SCOREZILLA_BETA_TOKEN— pre-public closed-beta only. When set, sent as theX-MCP-Betaheader on every API call to unlock the MCP namespace before the public switch is flipped. You'll only need this if a Scorezilla team member gave you a beta token; ignore otherwise.
Tokens: how they work
Tokens are scoped to the developer who issued them and see every game associated with their account.
The MCP server never returns the secret-key plaintext for a game — for that, copy from the dashboard.
Revoke a token any time at dashboard.scorezilla.dev/account/tokens. Revocations propagate within a few seconds.
Tokens are bearer credentials: anyone with the value can call the API on your behalf. Don't commit them to source; don't paste them into shared chats. Keep them in
envblocks, password managers, or secret stores.
Runtime requirements
Node ≥ 20
A network path to
https://api.scorezilla.dev
Releasing
Releases are CI-driven and require an approval click in the npm-publish GitHub Environment. The full flow:
Author a changeset locally:
pnpm changeset— describes what changed and the bump type. Commit the file under.changeset/.Merge to main.
.github/workflows/release.ymlruns and opens a "chore(release): version @scorezilla/mcp" PR that bumpspackage.json, syncsserver.json(the MCP Registry manifest) viascripts/sync-server-json-version.mjs, and updatesCHANGELOG.md.Merge the version PR. The same workflow then publishes:
npm tarball with
--provenance(verifiable build attestation via GH OIDC + sigstore)MCP Registry record via
mcp-publisher login github-oidc→mcp-publisher publishPost-publish smoke test that installs the published tarball and runs the binary
Pre-flight guards that run before publish: typecheck, test, build, bin smoke (
node dist/index.js --version), andrelease:check(assertspackage.jsonandserver.jsonversions agree).
Manual publishes from a developer terminal still work (bash scripts/publish.sh) but aren't the path CI takes — they skip provenance and approval gates. Use only for one-off recovery.
Issues / feedback
License
MIT.
Maintenance
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/isco-tec/scorezilla-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server