MCP YouTube Transcript
An MCP server retrieving transcripts of YouTube videos
作者提供的说明(英文原文)
MCP server for Argo RPG Platform — connects AI assistants to campaign data via OAuth2
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。
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。
选择你使用的 AI 应用,然后复制设置。
OAUTH_TOKEN, REFRESH_TOKEN, ARGO_API_BASE标为机密的值是你账户的钥匙。千万不要分享给别人,也不要粘贴到聊天里。
你需要
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 …”。
适用于网页版、桌面版和手机版 Claude:
https://mcp.argo.games/mcp打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json
{
"mcpServers": {
"argo": {
"command": "npx",
"args": [
"-y",
"argo-mcp"
],
"env": {
"OAUTH_TOKEN": "your-value",
"REFRESH_TOKEN": "your-value",
"ARGO_API_BASE": "your-value"
}
}
}
}在终端中运行这条命令
claude mcp add --transport http argo https://mcp.argo.games/mcpclaude mcp add argo --env OAUTH_TOKEN=your-value --env REFRESH_TOKEN=your-value --env ARGO_API_BASE=your-value -- npx -y argo-mcp运行前请先填写空着的值。
粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"argo": {
"url": "https://mcp.argo.games/mcp"
}
}
}粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"argo": {
"command": "npx",
"args": [
"-y",
"argo-mcp"
],
"env": {
"OAUTH_TOKEN": "your-value",
"REFRESH_TOKEN": "your-value",
"ARGO_API_BASE": "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。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"argo": {
"type": "http",
"url": "https://mcp.argo.games/mcp"
}
}
}粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"argo": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"argo-mcp"
],
"env": {
"OAUTH_TOKEN": "your-value",
"REFRESH_TOKEN": "your-value",
"ARGO_API_BASE": "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。
ChatGPT 只能连接互联网上的服务器:
https://mcp.argo.games/mcp菜单名称可能因套餐和版本不同而略有差异。
你的选择通过 Cookie 保存在这台设备上,无需账户。
ARGO_API_BASE env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。
我们还没有读取这个服务器的工具列表;我们会轮流检查各个端点。
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7日
注册名称已通过 GitHub 账户 Argo-RPG-Platform 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
尚未检查;我们会轮流检查各个端点
尚未检查;我们会轮流检查各个端点
尚未检查;我们会在检查端点时读取工具列表
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。
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。
你需要填写:OAUTH_TOKEN, REFRESH_TOKEN
两种方式都可以:安装在你的电脑上,或者连接云端版本。
GitHub 仓库最后更新于–,最新版本(1.0.3)于6个月前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。它通过 URL 连接,所以接受远程 MCP 服务器的应用也可以使用,例如开启开发者模式的 ChatGPT。
同一分类中有:MCP YouTube Transcript, Google Analytics MCP, BibiGPT YouTube Video Summarizer
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ 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
An MCP server retrieving transcripts of YouTube videos
GA4 MCP server for AI agents: schema discovery, server-side aggregation, analysis-ready defaults.
AI video summarizer: summarize and transcribe YouTube, Bilibili, TikTok and podcast videos.
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。
Connects AI assistants to your Argo campaigns via the Model Context Protocol. Once configured, your AI assistant can read and write campaign lore, look up character details, and interact with Argo data directly from the chat interface.
Supported clients: Claude Code, Claude Desktop, OpenAI Codex, VS Code (1.103+), and any MCP-compatible host.
Clicking either button hands a vscode:mcp/install URL to your editor with this config:
{ "type": "http", "url": "https://mcp.argo.games/mcp" }
VS Code stores the entry in your user-level mcp.json, then negotiates OAuth on first connect via Dynamic Client Registration against the Argo authorization server. No tokens are embedded in the URL — VS Code stores the resulting access and refresh tokens in your OS keychain.
Linux / Flatpak / Snap note: If clicking the button doesn't open VS Code, the vscode: URL handler isn't registered. This is common on:
update-desktop-database. Either run xdg-mime default code.desktop x-scheme-handler/vscode once, or use the manual fallback below.Manual fallback: copy the JSON from the VS Code (one-click, recommended) section below into your VS Code mcp.json config.
The same install button is rendered on the Argo install page at https://app.argo.games/docs/mcp/install.
There are three supported ways to use the Argo MCP server. Pick the one that matches your client.
Use the hosted endpoint — do not install the npx package. Add this MCP server URL to ChatGPT:
https://mcp.argo.games/mcp
ChatGPT will walk you through OAuth on first connect via dynamic client registration.
Sign in once on the machine that runs the MCP client:
npx -y argo-mcp auth login
The command prints the consent URL, opens nothing, and asks you to paste the access token (and optionally the refresh token) from the consent page. Tokens are saved locally:
%APPDATA%\argo-mcp\tokens.json~/Library/Application Support/argo-mcp/tokens.json${XDG_CONFIG_HOME:-~/.config}/argo-mcp/tokens.jsonThen configure the MCP server with no env vars:
{
"mcpServers": { "argo": { "command": "npx", "args": ["-y", "argo-mcp"] } } }
Other auth commands:
npx -y argo-mcp auth status # show whether you are signed in
npx -y argo-mcp auth logout # forget locally stored tokens
Local stdio mode does not use OAuth Dynamic Client Registration. DCR is only used by the hosted HTTP server at
https://mcp.argo.games.
For CI, Docker, or anyone who prefers explicit env config, you can skip auth login and pass tokens directly. OAUTH_TOKEN always wins over the local token store.
{
"mcpServers": { "argo": { "command": "npx", "args": ["-y", "argo-mcp"], "env": { "OAUTH_TOKEN": " ", "REFRESH_TOKEN": " " } } } }
git clone https://github.com/Argo-RPG-Platform/MCP.git argo-mcp
cd argo-mcp npm install npm run build
The MCP server authenticates with Argo using an OAuth2 access token tied to a specific set of campaigns. Tokens are obtained through the Argo consent flow:
OAUTH_TOKENREFRESH_TOKEN (recommended; enables automatic renewal when the access token expires after ~1 hour)Treat both tokens like passwords. Revoke them at any time from the campaign's integrations page.
Use the install buttons at the top of this README, or paste the following into a .vscode/mcp.json (per-workspace) or your user-level MCP config:
{
"servers": { "argo": { "type": "http", "url": "https://mcp.argo.games/mcp" } } }
VS Code handles OAuth automatically via Dynamic Client Registration on first connect. There is no need to set OAUTH_TOKEN or REFRESH_TOKEN — VS Code stores tokens in the OS keychain.
Add the server with the CLI:
claude mcp add argo \
--command node \ --args /absolute/path/to/argo-mcp/dist/index.js \ --env OAUTH_TOKEN= \ --env REFRESH_TOKEN=
Or add it manually to your project's .claude/settings.json:
{
"mcpServers": { "argo": { "command": "node", "args": ["/absolute/path/to/argo-mcp/dist/index.js"], "env": { "OAUTH_TOKEN": " ", "REFRESH_TOKEN": " " } } } }
Edit your Claude Desktop config file:
~/Library/Application Support/Claude/claude_desktop_config.json%APPDATA%\Claude\claude_desktop_config.json{
"mcpServers": { "argo": { "command": "node", "args": ["/absolute/path/to/argo-mcp/dist/index.js"], "env": { "OAUTH_TOKEN": " ", "REFRESH_TOKEN": " " } } } }
Restart Claude Desktop after saving.
Add the server to your Codex MCP configuration:
{
"mcpServers": { "argo": { "command": "node", "args": ["/absolute/path/to/argo-mcp/dist/index.js"], "env": { "OAUTH_TOKEN": " ", "REFRESH_TOKEN": " " } } } }
For the hosted ChatGPT app surface, this repo exposes OAuth discovery, dynamic client registration, and domain verification endpoints on the MCP domain. The submission checklist and required env vars are in CHATGPT_APP_SUBMISSION.md.
.env file (local dev)Create a .env file in the argo-mcp directory:
OAUTH_TOKEN=<your-access-token>
REFRESH_TOKEN=
Then run directly:
npm run dev # hot-reload via tsx
npm start # compiled dist/
Once the MCP is connected, you can install the argo-agent-skills plugin to get ready-made Game Master workflows that orchestrate the tools below — session prep, post-session recap, NPC generation, and a guided MCP setup walkthrough. The same skill source ships parallel packaging for Claude (Code + Claude.ai) and OpenAI-compatible clients (Codex, ChatGPT).
Claude Code:
/plugin marketplace add Argo-RPG-Platform/argo-agent-skills
/plugin install argo-gm-pack@argo
Claude.ai / Codex / ChatGPT: download the per-target zip from the latest release and follow the upload steps in that repo's README.
See the Argo Skills docs for the full list of skills, examples, and authoring guidance for your own.
This server exposes tools across campaigns, mnemons, sessions, guilds, friends, and the Argo community forum.
list_campaigns (campaign.read) lists accessible campaigns and their access levelget_campaign (campaign.read) retrieves campaign detailscreate_campaign (campaign.create) creates a new campaignupdate_campaign (campaign.write) updates campaign name or descriptionlist_co_gms (campaign.read) lists assistant GMsadd_co_gm (campaign.write) adds an assistant GMremove_co_gm (campaign.write) removes an assistant GMdescribe_mnemon_types lists supported mnemon types, relationship labels, and content-block ruleslist_mnemons (campaign.read) lists mnemon entries for a campaignget_mnemon (campaign.read) retrieves a mnemon entry in fulllist_mnemon_relationships (campaign.read) lists linked entries and relationship edgescreate_*_mnemons (campaign.write) creates typed mnemon entries for NPC, Location, Quest, Lore, Archive, Journal, SessionSummary, Player, and Custom entriesupdate_*_mnemons (campaign.write) updates typed/meta fields for those mnemon entry types; the quest tools also write a quest's objectives, reward chips, background and GM notesset_quest_objective_state (campaign.write) ticks, fails or reopens one quest objective by its idupdate_mnemons_content (campaign.write) edits mnemon content blockscreate_mnemon_relationship (campaign.write) creates a relationship between entriesdelete_mnemon_relationship (campaign.write) deletes a relationship by IDcreate_session (campaign.write) schedules a campaign sessionlist_sessions (campaign.read) lists sessions for a campaign and monthget_session (campaign.read) retrieves a single sessionupdate_session (campaign.write) updates a session's schedule or textlist_guilds (guild.read) lists guilds the current user belongs toget_guild (guild.read) retrieves guild detailslist_guild_members (guild.read) lists guild membersadd_campaign_to_guild (guild.write) links a campaign to a guildinvite_guild_member (guild.admin) invites a guild memberremove_guild_member (guild.admin) removes a guild memberset_guild_member_role (guild.admin) changes a guild member's roleadd_guild_calendar_event (guild.admin) adds a guild calendar eventlist_friends (friends.read) lists accepted friendslist_sent_friend_requests (friends.read) lists outgoing pending requestslist_received_friend_requests (friends.read) lists incoming pending requestssend_friend_request (friends.write) sends a friend requestaccept_friend_request (friends.write) accepts a requestreject_friend_request (friends.write) rejects a requestcancel_friend_request (friends.write) cancels a sent requestTo bring someone into a campaign or guild, use invite_guild_member or send_friend_request — the MCP surface does not send email to addresses outside your existing Argo context.
forum_list_categories (forum.read) lists forum categoriesforum_list_topics只显示了开头部分;其余内容请在仓库中阅读