原型 · 预览版:收录官方注册中心里的全部服务器,每 30 分钟同步一次。评价和账户功能将在后续阶段推出。
PopNx MCP为使用 AI 的人准备的 MCP 服务器 · 用大白话讲清楚

MCP Metacritic

作者:smeet666版本 2.1.0i★ –i在你的电脑上运行
注册名称io.github.smeet666/mcp-metacritic

安装

概览

作者提供的说明(英文原文)

Search Metacritic films, shows and games, read scores and critic reviews. No API key.

PopNx MCP 的三大支柱

这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。

它能帮你做什么

我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。

  • 连接方式: 在你的电脑上运行
  • 传输方式: stdio
  • 分类: 游戏、体育、美食与学习

安装

选择你使用的 AI 应用,然后复制设置。

安装之前

  • 基础检查:1 项中通过 1 项
  • 在你的电脑上运行,拥有你的用户账户的权限(文件、网络、应用)。
  • 需要你自己填写: MC_USER_AGENT, MC_MIN_INTERVAL_MS, MC_LOG_LEVEL

你需要

  • npm已安装 Node.js(用于 npx 命令)

你用的是哪个应用?

在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。

下载作者的 .mcpb 文件

链接指向作者自己的文件;我们不托管也不重新打包它。打开文件后,Claude Desktop 会进行安装。

在终端中运行这条命令

claude mcp add metacritic --env MC_USER_AGENT=your-value --env MC_MIN_INTERVAL_MS=your-value --env MC_LOG_LEVEL=your-value -- npx -y mcp-metacritic

运行前请先填写空着的值。

在 Cursor 中打开

或者自己复制设置

粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json

{
  "mcpServers": {
    "metacritic": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-metacritic"
      ],
      "env": {
        "MC_USER_AGENT": "your-value",
        "MC_MIN_INTERVAL_MS": "your-value",
        "MC_LOG_LEVEL": "your-value"
      }
    }
  }
}

浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。

在 VS Code 中打开

或者自己复制设置

粘贴到项目中的 .vscode/mcp.json

{
  "servers": {
    "metacritic": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "mcp-metacritic"
      ],
      "env": {
        "MC_USER_AGENT": "your-value",
        "MC_MIN_INTERVAL_MS": "your-value",
        "MC_LOG_LEVEL": "your-value"
      }
    }
  }
}

浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。

这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。

你的选择通过 Cookie 保存在这台设备上,无需账户。

需要你填写的值

  • MC_USER_AGENT env
    Override the User-Agent sent to Metacritic.
  • MC_MIN_INTERVAL_MS env
    Minimum milliseconds between requests. Default 1000, floor 500.
  • MC_LOG_LEVEL env
    silent, error, info or debug. Default error.

这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。

它能做什么

它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。

基础检查i

我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月8日

  • ✓ 通过

    作者与仓库一致

    注册名称已通过 GitHub 账户 smeet666 验证,该账户拥有此仓库

  • … 尚未检查

    仍在维护

    我们还没有从 GitHub 读取这个仓库

  • … 尚未检查

    有许可证

    我们还没有从 GitHub 读取这个仓库

  • – 不适用

    端点有响应

    它在你的电脑上运行,没有互联网端点

  • – 不适用

    使用 OAuth 登录

    在你的电脑上运行,改用本地设置

  • … 尚未检查

    工具描述中没有隐藏指令

    它在你的电脑上运行;我们只能读取互联网端点的工具列表

与“游戏、体育、美食与学习”中的其他服务器比较i

  • 星标数多于此分类中其他 1301 个服务器的 0%
  • 发布到注册中心的时间比此分类中其他 1301 个服务器的 76% 更新
  • 基础检查结果等于或优于此分类中其他 1301 个服务器的 100%

与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。

常见问题

MCP Metacritic 安全吗?

它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。

它需要哪些密钥或代码?

注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。

它在我的电脑上运行还是在云端?

它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。

它还在维护吗?

GitHub 仓库最后更新于–,最新版本(2.1.0)于10天前进入注册中心。

哪些应用可以使用它?

支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js, MCPB。

有什么可以替代它?

同一分类中有:Convoca, mcp.film directory, 像塑 MCP Plus

替代选择

同一分类中其他作者的服务器,按星标数排序。

如何看卡片:↓ 每周下载量i★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i

Convoca

Open literary prizes and contests in Spanish and Catalan, with winners and awarded books.

作者:convocaliteraria.com起步阶段今天更新

在云端运行✓ convocaliteraria.com

mcp.film directory

Search the curated directory of MCP servers for AI filmmaking and plan a full production stack.

作者:c47-inc起步阶段↓ 113/周今天更新

电脑或云端✓ mcp.film

像塑 MCP Plus

Unofficial macOS MCP for Douyin AR (像塑): assets, menus, AI combine, delivery. For learning only.

作者:Muqian-Sun非常新今天更新

在你的电脑上运行

作者的 README

打开 README

按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。

mcp-metacritic

Metacritic gathers what critics and audiences said about films, television series and video games. Each entry carries the year, the age rating, the genres, and two scores of its own: the Metascore, a weighted average of the professional reviews, and the user score, out of ten, from the people who signed up to rate it. Under each entry sit the reviews themselves, with the publication that ran them and the quoted line.

This server connects a chat client to that catalogue. You can search for a title, read its entry with its scores and its details, browse a catalogue by score, recency or popularity, and read the reviews of one title, filtered by critic or audience and by how favourable they were. It needs no API key and no account.

_Version française_


Install

One-click install

Claude Code

claude mcp add metacritic -- npx -y mcp-metacritic

Claude Desktop, Cursor, and any client using the standard config format

{

"mcpServers": { "metacritic": { "command": "npx", "args": ["-y", "mcp-metacritic"] } } }

Node 24 or later is required, and no environment variable has to be set.

With Docker
{

"mcpServers": { "metacritic": { "command": "docker", "args": ["run", "-i", "--rm", "ghcr.io/smeet666/mcp-metacritic:2.1.0"] } } }

-i keeps stdin open, which is where the protocol travels, and -t is left out because a TTY rewrites the stream. The container needs outbound HTTPS to backend.metacritic.com, and nothing else: no volume, no port, no credential.

Bundle, without npm

Download mcp-metacritic-2.1.0.mcpb from the latest release and open it. A client that supports MCP bundles installs it on its own, with no npm and no configuration file to edit. The bundle carries its dependencies, so nothing is fetched at install time.

What you can ask

  • "What did critics make of The Matrix?"
  • "Read me a few negative reviews of that game."
  • "What are the best-reviewed horror films?"
  • "How does the user score compare to the Metascore?"
  • "What came out recently that reviewed well?"

The ordinary path runs from a search to an entry: a row carries a slug and a kind, and get_title and get_reviews take both together.

Tools

ToolWhat it does
search_titlesFinds films, series and games by title.
get_titleReads one entry, its scores and its details.
get_reviewsReads the reviews of one entry, by source and by sentiment.
browse_titlesLists a catalogue by score, recency or popularity.

A title is addressed by its slug together with its kind, since the same slug can name a film and a game.

search_titles

Finds films, series and games by title.

ArgumentTypeRequiredWhat it does
querystring, at least 1 characteryesA title, or part of one.
kindmovie, show, game or any, default anynoWhich catalogue to search.
limitinteger, 1 to 50, default 10noRows to serve.

In return: rows carrying slug and kind, which get_title and get_reviews take together; title; year; release_date; rating, the age rating as published; metascore; user_score, which the search route publishes for no row and which is null on every one of them; and source_url. A score the site has not computed is `null`, never `0`: on a scale that starts at zero the two would be indistinguishable, and a title with too few reviews carries none.

get_title

Reads one entry. The heavier parts are asked for rather than served by default, and each one beyond the default costs a request.

ArgumentTypeRequiredWhat it does
slugstring, at least 1 characteryesThe identifier a row carries.
kindmovie, show or gameyesWhich catalogue it belongs to.
sectionsarray of basic, scores, awards, production, networks, where_to_watch, default ["basic", "scores"]noWhich parts to return.
max_charsinteger, 200 to 20000, default 4000noHow much of the description to serve.
offsetinteger, 0 or more, default 0noWhere to resume the description.

In return: the entry a search row carries, plus description, tagline, genres, duration_minutes and imdb_id, each null where the page states nothing. total_chars, returned_chars and offset describe the slice of the description served.

get_reviews

Reads the reviews of one entry.

ArgumentTypeRequiredWhat it does
slugstring, at least 1 characteryesThe identifier a row carries.
kindmovie, show or gameyesWhich catalogue it belongs to.
sourcecritic or user, default criticnoWhose reviews to read.
sentimentall, positive, neutral or negative, default allnoHow favourable a review has to be.
limitinteger, 1 to 50, default 10noReviews to serve.
offsetinteger, 0 or more, default 0noReviews to skip, for paging.

In return: reviews, each with its quote as published, its score, the max that score is out of, which is 100 for a critic and 10 for a user, and the publication that ran it. Name the publication when quoting a review. total_available counts every review Metacritic holds from that source for the entry, whatever sentiment was asked for, since the sentiment sorts the sample this server received rather than the request that fetched it. next_offset continues.

browse_titles

Lists a catalogue.

ArgumentTypeRequiredWhat it does
kindmovie, show or game, default movienoWhich catalogue to list.
sortscore, recent or popular, default scorenoHow the rows are ordered.
genrestringnoA single genre name, such as Horror.
limitinteger, 1 to 50, default 20noRows to serve.
offsetinteger, 0 or more, default 0noRows to skip, for paging.

In return: the rows search_titles returns, with total_available, offset, next_offset and the kind, sort and genre the listing was read under.

Two scores, two things measured

The Metascore is a weighted average of professional reviews, out of 100. The user score is the average of what registered members gave, out of 10. They measure different populations on different scales, and a title can carry one and not the other. Read each with the max its reviews state, and report a missing score as missing.

Configuration

Every variable is optional. Set them in the env block of your client config.

VariableDefaultWhat it does
MC_USER_AGENTthe project identityNames your application to the site, with an address where a person can be reached.
MC_MIN_INTERVAL_MS1000Gap between two requests, from 500 to 60000.
MC_TIMEOUT_MS15000Deadline for one request, from 1000 to 120000.
MC_MAX_RETRIES3Attempts after a transient failure, from 0 to 10.
MC_CACHE_TTL_MS86400000How long a catalogue entry stays in memory, from 0 to 604800000.
MC_SCORES_CACHE_TTL_MS3600000How long scores and reviews stay in memory, from 0 to 86400000.
MC_CACHE_MAX_ENTRIES200Answers held in memory at once, from 0 to 10000.

| MC_LOG_LEVEL | error

只显示了开头部分;其余内容请在仓库中阅读

问 PopNx