
Meltbead Craft Colors
io.github.jsspace由 GitHub 帳號 jsspace 發布v0.1.0更新於 Nov 3, 2026
使用 Meltbead 搜尋手工配色方案,並按 CIEDE2000 符合色號。
AI 翻譯由 AI 模型根據下方的原文描述翻譯,可能有誤。
原文描述
Search craft color palettes and match shades by CIEDE2000 with Meltbead.
發布方自己的服務:Craft
概覽
Meltbead Craft Colors 的本機 MCP 服務;無需登入;STDIO。
暫無 AI 概覽:以下資訊來自註冊中心條目和服務本身。
安裝
在 SourceWeft 中
- 開啟 儀表板中的 Meltbead Craft Colors,將其新增到工作區。
- 為需要使用其工具的對話啟用該服務。
Desktop only,透過 STDIO。 STDIO 服務會啟動本機處理程序,因此需要 SourceWeft 桌面主機。
其他 MCP 客戶端
參照 儲存庫 中的啟動說明。
README
Craft Colors MCP
A local MCP server for craft color charts and approximate screen-color matching. It combines the existing @meltbead/bead-colors fuse bead references with eight locally bundled palettes distributed by Meltbead under direct authorization. No account or network request is needed after installation.
Requirements
Node.js 20 or newer.
Install from npm
The server uses stdio: an MCP client launches it and communicates through stdin/stdout. Starting it in a terminal will appear idle until a client connects. Add this configuration to an MCP client:
Package: @meltbead/bead-colors-mcp on npm.
Tools
The service exposes 19 palettes: 11 existing fuse bead palettes plus Miyuki Delica, DMC floss, DMC-coded diamond drills, Red Heart yarn, LEGO, Minecraft blocks, Minecraft map colors, and a generated pixel-art palette. Search and cross-palette matching use representative palettes by default; subset palettes and the generated palette remain accessible by ID. Matching excludes effect finishes unless includeEffects is true. Discontinued Delica colors and unbuildable Minecraft map colors are excluded from automatic matches but remain available by code. Transparent entries without one RGB value cannot be matched.
The eight bundled palettes are available as JSON in data/palettes/ and CSV in data/csv/. MARD retail-box membership is also available as data/csv/mard-kits.csv. CSV rows include Meltbead attribution, source URL, license, and status. Run npm run build:csv after editing JSON, then npm run check:csv to verify the exports.
The matching score is CIEDE2000 (deltaE), calculated from each palette's sRGB screen reference. Smaller scores mean closer screen colors. They are not physical measurements or guarantees of material equivalence. get_color accepts case-insensitive codes and common separators; Delica codes also accept shorthand such as DB10, DB-010, and 10 for DB0010.
Every color result includes a reminder to verify important shades against physical materials and printed product codes. Palette metadata includes the original data source and its status where supplied.
Develop
npm start runs the local source server. smoke starts it through a real MCP stdio client, lists its tools, and calls them.
License and data sources
The server code is MIT licensed. The files in data/ are published under CC BY 4.0, with Meltbead as the attribution party. Suggested credit: Color data: Meltbead (https://meltbead.com/). The additional palettes and MARD retail-box data are distributed under separate authorization from their rights holder. Underlying references and quality notes are in data/palettes/ and DATA_SOURCES.md. The existing @meltbead/bead-colors dependency retains its third-party notices. Brand names belong to their owners.
來源:README.md,提交 22963e1
工具
0版本歷史
1- v0.1.0最新Nov 3, 2026
評分
已滿足適用檢查項的 50%
該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。
- 近 30 天內可連通25 分
請保持端點能夠回應 MCP 握手;SourceWeft 每 30 天探測一次。
- 已宣告認證方式10 分
請在註冊中心條目中宣告憑證請求標頭,或保持端點能夠回應探測,以便判定其認證方式。
- 已列出工具15 分
請讓伺服器在無憑證時回應 tools/list,或發布可在沙箱中無憑證啟動的軟體包。
- 描述必需15 分
- 圖示10 分
- 文件15 分
- 授權條款10 分
其他 Craft 整合
Craft
do.craft.mcp官方發布者
Craft 的託管 MCP 服務,由 mcp.craft.do 發布;需要憑證;Streamable HTTP。

Agent Craft
由 GitHub 帳號 CipherClaw 發布
一個只有AI代理程式才能遊玩的持久體素生存世界:挖礦、製作、建造、交易、聊天。AI 翻譯

SonicCraft
由 GitHub 帳號 user-github-me 發布
把 YAML/JSON 音訊時間軸編譯成 TTS 語音、音樂與音效,並進行閃避、響度母帶處理與對應字幕。

SlideCraft
由 GitHub 帳號 eiler2005 發布
SlideCraft 把一份簡報需求變成可編輯的 PowerPoint、PDF、HTML 與 Word 簡報,並附排版適配、來源查核與視覺審閱。

CoverCraft AI by Ambuj Kumar Tripathi
由 GitHub 帳號 Ambuj123-lab 發布
求職導向的 MCP 伺服器:調查目標公司、以網路、GitHub 與 Hugging Face 證據核對履歷說法,並產生附引用的求職信。