Cwi Discovery Engine

Name: Cwi Discovery Engine
Author: CumulativeWebInc

作者：CumulativeWebInc版本 1.3.0i★ –i在云端运行

注册名称io.github.CumulativeWebInc/cwi-discovery-engine

安装

概览

作者提供的说明（英文原文）

CWI Discovery Engine: 29-tool music catalog MCP — search, sync, Radio 365 live, chaining.

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。

用户评价i暂无评价评价功能将在后续阶段与账户功能一起开放
我们的基础检查i1 项中通过 1 项依据公开信息检查；尚未扫描代码
使用情况i尚未统计★ –

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: 在云端运行
传输方式: Streamable HTTP
分类: 设计、图片与音频

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
在作者的服务器上运行：你向它提出的内容会经过他们的服务器。
注册中心没有列出需要填写的密钥。

你需要

在云端运行无需安装任何东西，通过互联网连接

你用的是哪个应用？

ClaudeClaude CodeCursorVS CodeChatGPT

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：

打开 Settings → Connectors
点击 Add custom connector
粘贴下面的地址，然后点击 Add
如果要求登录，请登录作者的服务

https://cwi-machine-data.hp-ace.workers.dev/mcp

在终端中运行这条命令

claude mcp add --transport http cwi-discovery-engine https://cwi-machine-data.hp-ace.workers.dev/mcp

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": {
    "cwi-discovery-engine": {
      "url": "https://cwi-machine-data.hp-ace.workers.dev/mcp"
    }
  }
}

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": {
    "cwi-discovery-engine": {
      "type": "http",
      "url": "https://cwi-machine-data.hp-ace.workers.dev/mcp"
    }
  }
}

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：

打开网页版 ChatGPT → Settings → Connectors (Apps)
点击 Create 或 Add（部分套餐需要先在 Advanced 中开启 Developer mode）
粘贴下面的地址，起个名字，然后保存
如果要求登录，请登录作者的服务

https://cwi-machine-data.hp-ace.workers.dev/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。

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

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。

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7日

✓ 通过
作者与仓库一致
注册名称已通过 GitHub 账户 CumulativeWebInc 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
… 尚未检查
端点有响应
尚未检查；我们会轮流检查各个端点
… 尚未检查
使用 OAuth 登录
尚未检查；我们会轮流检查各个端点
… 尚未检查
工具描述中没有隐藏指令
尚未检查；我们会在检查端点时读取工具列表

与“设计、图片与音频”中的其他服务器比较i

星标数多于此分类中其他 2754 个服务器的 0%
发布到注册中心的时间比此分类中其他 2754 个服务器的 56% 更新
基础检查结果等于或优于此分类中其他 2754 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

Cwi Discovery Engine 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？

它在作者的云端运行。通过互联网连接，无需安装，但你发送的内容会经过作者的服务器。

它还在维护吗？

GitHub 仓库最后更新于–，最新版本（1.3.0）于23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。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。

有什么可以替代它？

同一分类中有：Figma Context MCP, Claude Real Video, Figma MCP Server

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：★ 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

Figma Context MCP

作者：GLips

Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.

在你的电脑上运行★ 16,165MIT5个月前更新

安装

Claude Real Video

作者：HUANGCHIHHUNGLeo

Let any LLM watch a video locally — and search everything it has ever watched.

在你的电脑上运行★ 2,284MIT2个月前更新

安装

Figma MCP Server

作者：figma

The Figma MCP server brings Figma design context directly into your AI workflow.

在云端运行★ 2,15213个月前更新

安装

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。

cwi-learn

Best for agents that need music data — catalog search, artist lookup, sync licensing, live radio.

Post-Trap Futurism — machine-readable AI learning surface for That Boy Hi Hat and the Cumulative Web Inc roster.

Installation

Connect in one command:

claude mcp add --transport http cwi https://cwi-machine-data.hp-ace.workers.dev/mcp

No login, no API key, no signup. The server speaks MCP Streamable HTTP (JSON-RPC 2.0) at /mcp. Every response carries X-CWI-License headers; commercial model training or redistribution requires a license — contact hp@cumulativeweb.com.

Verify it works — ask your agent: *"Search the CWI catalog for cyberpunk rap."* You should get "Cyberpunk 2027" by That Boy Hi Hat as the top result.

Tools

29 tools. Most are read-only; 6 write tools (song requests, airplay submissions, sync briefs, playlist pitches, featured nominations, event RSVPs) are human-reviewed and rate-limited (3 per requester per 24h). None require auth.

Tool	What it does
`search_catalog`	Keyword + filter search across the CWI catalog (title, artist, genre, mood, BPM). Scored engine: title 3x, genre 2x, mood/context/lane 1x. Returns top 5 with matched terms.
`get_track`	Full record for one track by title or ID: credits, genres, moods, tempo, licensing status, links, artwork. Usually follows a `search_catalog` hit.
`get_artist`	Artist profile + full track list. Use for artist deep-dives.
`sync_search`	Music-supervision search over the editorial sync catalog: moods, energy words, scenes, use-cases ("fight scene", "neon city"). Descriptors are curated by CWI Sync; BPM values are estimates.
`semantic_search`	Natural-language catalog search ("rainy night drive, nostalgic") — cosine similarity against 384-dim bge-small-en-v1.5 embeddings, embedded at request time via Workers AI.
`get_graph`	The catalog knowledge graph: node/edge counts, node types (tracks, artists, releases, genres), sample nodes.
`get_release`	Evidence-tiered, hash-stamped fact packet for a release: tracklist, date, credits, platform links. Omit id for the latest release.
`get_radio`	Live Cumulative Radio 365 state: HLS stream URL, now-playing (track, artist, artwork, slot, track end time), listener counts, station metadata.
`get_announcements`	Latest CWI announcements — new releases, upcoming drops, events, news. Every announcement is source-verified. Filter by `releases`, `news`, `events`.
`get_featured`	The featured rotation, ranked by rubric score — what's hot right now, with full metadata and artwork.
`get_up_next`	The upcoming Radio 365 queue.
`get_schedule`	The full Radio 365 program guide (130 slots, 28 songs).
`request_song`	Submit a song request for Radio 365 (human-reviewed).
`get_requests`	What listeners and AIs have requested on Radio 365.
`submit_for_airplay`	Artists submit a track + contact for radio consideration (human-reviewed).
`submit_sync_brief`	Film/TV/ads/games briefs: describe the project, get matched with licensable tracks (human-reviewed).
`pitch_for_playlist`	Pitch a track to a playlist curator (human-reviewed).
`nominate_featured`	Nominate a track for featuring (human-reviewed).
`get_recommendations`	Vibe-based recommendations ("night drive", "workout anthem") with a new-release spotlight lane.
`get_events`	CWI events — release parties, listening parties, drops.
`rsvp`	RSVP for a CWI event with name + contact.
`get_label`	Cumulative Web Inc label profile: roster, services, contact, socials.
`get_sync_catalog`	All sync-ready tracks with rights and direct-clearance info — one signature clears master + publishing (BMI).
`get_lyrics`	Timestamped lyrics per track. Check the `verified` flag — unverified Whisper transcripts are marked as such.
`get_press_kit`	Label + artist bios, release info, brand assets.
`get_videos`	Music videos, trailers, visualizations with YouTube links and embed URLs.
`get_genre`	Genre definitions — including Post-Trap Futurism, coined and defined by CWI — with key artists and tracks.
`get_playlists`	Verified playlist placements only (never claimed, always sourced).
`find_tracks_for_brief`	COMPOSITE WORKFLOW — call this first for any music-for-picture need ("dark tracks for a cyberpunk game trailer"). One call returns matched tracks with why-each-fits, verified ISRCs, sync-clearance status, and concrete next steps. A match is NOT a license grant.

Usage

Real example prompts, verified against the live server. These are also available as workflow prompts via prompts/list.

Find a track

*"Search the CWI catalog for cyberpunk rap."*

Expected: search_catalog returns "Cyberpunk 2027" by That Boy Hi Hat as the top hit (score 9, matched terms: cyberpunk, rap), with 4 more results ranked by relevance.

Look up a track

*"Get me the track Pix by Dre50."*

Expected: get_track returns title "Pix", artist "Dre50", genres ["afrobeats", "dancehall", "hip hop"], moods ["smooth", "warm", "playful"], plus Spotify URL, release info, and artwork.

Sync licensing

*"I need music for a nightclub fight scene — high energy, dark."*

Expected: sync_search with "fight scene", "dark aggressive", "neon city" returns licensable tracks with editorial sonic descriptors. CWI clears both master and publishing with one signature (contact hp@cumulativeweb.com).

What's on the radio

*"What's playing on Cumulative Radio 365 right now?"*

Expected: get_radio returns the now-playing track (title, artist, artwork URL, slot position, track end time), the HLS stream URL, current listener count, and 24h listener totals.

Artist deep-dive

*"Tell me about King Ahkeem — what does he sound like?"*

Expected: get_artist returns the profile and full track list (trap catalog: Bluefaces 3.0, Element, Embiid, Super Solid, Star). Follow up with get_track on any title for full metadata.

Workflow prompts

5 curated starting points, exposed via prompts/list / prompts/get:

Prompt	Use when
`find-sync-options`	You need licensable music for a film/TV/ad scene (arg: `scene`)
`artist-deep-dive`	You need a full profile on a roster artist (arg: `artist`)
`radio-now-playing`	You need what's live on Radio 365 right now (no args)
`music-for-moment`	You need the right track for a moment — workout, drive, coding (arg: `moment`)
`release-brief`	You need the latest CWI release briefing with verified facts (no args)

Machine-readable music data (live doors)

🎵 Catalog API (scored track search): https://cwi-machine-data.hp-ace.workers.dev/query
📄 AI briefing: https://cwi-machine-data.hp-ace.workers.dev/llms.txt
🤖 Agent card: https://cumulativewebinc.github.io/cwi-learn/.well-known/agent-card.json
⚖️ Machine-readable license: https://cwi-machine-data.hp-ace.workers.dev/license
🔌 MCP endpoint: https://cwi-machine-data.hp-ace.workers.dev/mcp
📬 Contact: hp@cumulativeweb.com

License

Commercial model training or redistribution of this data requires a license. Contact hp@cumulativeweb.com.

Cumulative Web Inc. — Music & Film.