能力标签
OpenPencil开源向量设计工具
🔌
MCP工具

OpenPencil开源向量设计工具

基于 TypeScript · 让 AI 助手直接操作你的系统与工具
英文名:openpencil
⭐ 3.1k Stars 🍴 314 Forks 💻 TypeScript 📄 MIT 🏷 AI 8.5分
8.5AI 综合评分
向量设计AI原生MCP协议Agent协作开源设计工具
✦ AI Skill Hub 推荐

AI Skill Hub 强烈推荐:OpenPencil开源向量设计工具 是一款优质的MCP工具。已获得 3.1k 颗 GitHub Star,AI 综合评分 8.5 分,在同类工具中表现稳健。如果你正在寻找可靠的MCP工具解决方案,这是一个值得深入了解的选择。

📚 深度解析

OpenPencil开源向量设计工具 是一款基于 MCP(Model Context Protocol)标准协议的 AI 工具扩展。MCP 协议由 Anthropic 开发并开源,旨在建立 AI 模型与外部工具之间的标准化通信接口,目前已被 Claude Desktop、Claude Code、Cursor 等主流 AI 工具采纳。

通过安装 OpenPencil开源向量设计工具,你的 AI 助手将获得额外的工具调用能力,可以用自然语言直接操控该工具的功能,无需学习复杂的命令行语法。MCP 工具的核心价值在于"一次配置,永久增强"——配置完成后,每次与 AI 对话时都可以无缝调用这些工具。

在技术实现上,MCP 工具通过标准的 JSON-RPC 协议与 AI 客户端通信,工具的功能以"工具列表"的形式暴露给 AI 模型,AI 可以按需调用。OpenPencil开源向量设计工具 提供了结构化的工具调用接口,使 AI 模型能够精确地理解和使用每个功能点,显著降低 AI 在工具使用上的错误率。

与传统的 API 集成相比,MCP 工具的优势在于无需编写代码——用户只需在配置文件中添加几行 JSON,即可让 AI 获得全新能力。AI Skill Hub 将 OpenPencil开源向量设计工具 评为 AI 评分 8.5 分,属于同类工具中的优质选择。

📋 工具概览

全球首个开源AI原生向量设计工具,集成MCP协议支持AI Agent团队协作。基于TypeScript构建,提供AI辅助设计能力,适合设计师、开发者和AI应用开发者使用,推动设计自动化和智能化。

OpenPencil开源向量设计工具 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。

GitHub Stars
⭐ 3.1k
开发语言
TypeScript
支持平台
Windows / macOS / Linux
维护状态
持续维护,定期更新
开源协议
MIT
AI 综合评分
8.5 分
工具类型
MCP工具
Forks
314

📖 中文文档

以下内容由 AI Skill Hub 根据项目信息自动整理,如需查看完整原始文档请访问底部「原始来源」。

全球首个开源AI原生向量设计工具,集成MCP协议支持AI Agent团队协作。基于TypeScript构建,提供AI辅助设计能力,适合设计师、开发者和AI应用开发者使用,推动设计自动化和智能化。

OpenPencil开源向量设计工具 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。

📌 核心特色
  • 通过标准 MCP 协议与 Claude、Cursor 等主流 AI 客户端深度集成
  • 提供结构化工具调用接口,显著降低 AI 集成复杂度
  • 支持 Claude Desktop 和 Claude Code 无缝接入,开箱即用
  • 可与其他 MCP 工具组合叠加,构建完整 AI 工作站
  • 轻量无侵入设计,不影响现有系统架构
🎯 主要使用场景
  • 在 Claude Desktop 对话中直接调用本地工具,实现 AI 与系统的深度联动
  • 通过自然语言驱动复杂的多步骤自动化任务,代替繁琐手动操作
  • 将多个 MCP 工具组合使用,构建个人专属 AI 工作站
以下安装命令基于项目开发语言和类型自动生成,实际以官方 README 为准。
安装命令
# 方式一:通过 Claude Code CLI 一键安装
claude skill install https://github.com/ZSeven-W/openpencil

# 方式二:手动配置 claude_desktop_config.json
{
  "mcpServers": {
    "openpencil--------": {
      "command": "npx",
      "args": ["-y", "openpencil"]
    }
  }
}

# 配置文件位置
# macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
# Windows: %APPDATA%/Claude/claude_desktop_config.json
📋 安装步骤说明
  1. 确认已安装 Node.js(v18 或以上版本)
  2. 打开 Claude Desktop 或 Claude Code 的 MCP 配置文件
  3. 按「交给 Agent 安装 → Claude Desktop」标签中的 JSON 配置填入 mcpServers 字段
  4. 保存配置文件并重启 Claude 客户端
  5. 重启后,在对话中即可使用本工具
以下用法示例由 AI Skill Hub 整理,涵盖最常见的使用场景。
常用命令 / 代码示例
# 安装后在 Claude 对话中直接使用
# 示例:
用户: 请帮我用 OpenPencil开源向量设计工具 执行以下任务...
Claude: [自动调用 OpenPencil开源向量设计工具 MCP 工具处理请求]

# 查看可用工具列表
# 在 Claude 中输入:"列出所有可用的 MCP 工具"
以下配置示例基于典型使用场景生成,具体参数请参照官方文档调整。
配置示例
// claude_desktop_config.json 配置示例
{
  "mcpServers": {
    "openpencil________": {
      "command": "npx",
      "args": ["-y", "openpencil"],
      "env": {
        // "API_KEY": "your-api-key-here"
      }
    }
  }
}

// 保存后重启 Claude Desktop 生效
📑 README 深度解析 真实文档 完整度 64/100 查看 GitHub 原文 →
以下内容由系统直接从 GitHub README 解析整理,保留代码块、表格与列表结构。

简介

<p align="center"> <img src="./crates/op-host-desktop/assets/icon.png" alt="OpenPencil" width="120" /> </p>

OpenPencil

<p align="center"> <strong>The world's first open-source AI-native vector design tool.</strong><br /> <sub>Concurrent Agent Teams &bull; Design-as-Code &bull; Built-in MCP Server &bull; Multi-model Intelligence</sub> </p>

<p align="center"> <a href="./README.md"><b>English</b></a> · <a href="./README.zh.md">简体中文</a> · <a href="./README.zh-TW.md">繁體中文</a> · <a href="./README.ja.md">日本語</a> · <a href="./README.ko.md">한국어</a> · <a href="./README.fr.md">Français</a> · <a href="./README.es.md">Español</a> · <a href="./README.de.md">Deutsch</a> · <a href="./README.pt.md">Português</a> · <a href="./README.ru.md">Русский</a> · <a href="./README.hi.md">हिन्दी</a> · <a href="./README.tr.md">Türkçe</a> · <a href="./README.th.md">ไทย</a> · <a href="./README.vi.md">Tiếng Việt</a> · <a href="./README.id.md">Bahasa Indonesia</a> </p>

<p align="center"> <a href="https://github.com/ZSeven-W/openpencil/stargazers"><img src="https://img.shields.io/github/stars/ZSeven-W/openpencil?style=flat&color=cfb537" alt="Stars" /></a> <a href="https://github.com/ZSeven-W/openpencil/blob/main/LICENSE"><img src="https://img.shields.io/github/license/ZSeven-W/openpencil?color=64748b" alt="License" /></a> <a href="https://github.com/ZSeven-W/openpencil/actions/workflows/rust-check.yml"><img src="https://img.shields.io/github/actions/workflow/status/ZSeven-W/openpencil/rust-check.yml?label=CI" alt="CI" /></a> <a href="https://discord.gg/h9Fmyy6pVh"><img src="https://img.shields.io/badge/Discord-Join%20chat-5865F2?logo=discord&logoColor=white" alt="Discord" /></a> </p>

<p align="center"> <a href="https://trendshift.io/repositories/24088?utm_source=repository-badge&amp;utm_medium=badge&amp;utm_campaign=badge-repository-24088" target="_blank" rel="noopener noreferrer"><img src="https://trendshift.io/api/badge/repositories/24088" alt="ZSeven-W%2Fopenpencil | Trendshift" width="250" height="55" /></a> </p>

<br />

<p align="center"> <a href="https://oss.ioa.tech/zseven/openpencil/a46e24733239ce24de36702342201033.mp4"> <img src="./screenshot/op-cover.png" alt="OpenPencil — click to watch demo" width="100%" /> </a> </p> <p align="center"><sub>Click the image to watch the demo video</sub></p>

Features

Canvas & Drawing

  • Infinite canvas with pan, zoom, smart alignment guides, and snapping
  • Rectangle, Ellipse, Line, Polygon, Pen (Bezier), Frame, Text
  • Boolean operations — union, subtract, intersect with contextual toolbar
  • Icon picker (Iconify) and image import (PNG/JPEG/SVG/WebP/GIF)
  • Auto-layout — vertical/horizontal with gap, padding, justify, align
  • Multi-page documents with tab navigation

Design System

  • Design variables — color, number, string tokens with $variable references
  • Multi-theme support — multiple axes, each with variants (Light/Dark, Compact/Comfortable)
  • Component system — reusable components with instances and overrides
  • CSS sync — auto-generated custom properties, var(--name) in code output
  • Reusable UIKits — import/export component kits from .pen files

AI & Agents

  • Prompt-to-canvas with streaming generation and orchestrator-driven spatial decomposition
  • Concurrent Agent Teams — multiple designers work on different sections in parallel with per-member canvas indicators
  • Layered workflow — design_skeletondesign_contentdesign_refine with focused prompts per phase
  • Style Guides — 50+ built-in styles (glassmorphism, brutalist, retro, etc.) with tag-based fuzzy matching, wired into planning and generation
  • Multi-model capability profiles — auto-adapts thinking mode, effort, and prompt shape per model tier
  • Built-in Rust agent runtime + Anthropic, Claude Agent SDK, OpenCode, Codex, Copilot, Google Gemini API providers
  • Anthropic-format passthrough for Chinese LLM providers — Kimi, Zhipu, GLM, DouBao, Ark, Bailian/DashScope, ModelScope, Coding Plans

Collaboration

  • Authenticated peer-to-peer sessions with a public relay fallback and built-in regional collaboration hubs
  • Join with a 10-character region-tagged pairing code — no account required for LAN sessions
  • Live remote cursors, cross-account collaboration, and a conflict panel with per-edit detail and replay of discarded edits
  • Online multi-tenant mode for the --serve-web daemon, authenticated against the op-hub with cross-account tenant sharing
  • Device login — sign in from the browser through the serve-web daemon, with profile avatars and usernames in the editor

Presentation Decks

  • Six 16:9 deck templates with a template picker, plus AI deck planning at projector size — one slide per screen
  • Present a deck as a slideshow with presenter controls
  • Export a deck as PDF (one page per slide), a self-contained slideshow HTML file, an editable PowerPoint (.pptx), or a hyperframes video composition
  • Slides rail navigator; the agent validates deck board geometry (aspect, overflow, centering) as well as the prompt

Templates & Web Capture

  • Scene template center — a browsable catalogue of 58 templates across six scenes, opened from File ▸ New from template
  • Prompt center with web, dashboard, component, and modify entries and visual prompt previews
  • Asset center — a full-window responsive gallery with dual-action templates and DESIGN.md style import
  • Chrome web-capture extension — snapshot any webpage into OpenPencil with faithful HTML / layout capture; snapshots are delivered to your account's inbox on the hub, or downloaded as a ready-to-open .op file offline

Git Integration

  • Clone wizard with SSH / HTTPS auth and SSH key management
  • Branch picker — create, switch, delete, merge, all from the git panel
  • Pull / push cascades with auth retry and non-fast-forward handling
  • Folder-mode three-way merge with on-disk MERGE_HEAD state tracking
  • Conflict panel with per-node / per-field three-way cards, inline JSON editor, bulk actions, and inline diff block
  • Remote settings and SSH keys UI; 15-locale i18n across the whole Git surface

Export

  • Canvas export — PNG, JPEG, WEBP, PDF (Cmd+Shift+P)
  • Code export — React + Tailwind, HTML + CSS, Vue, Svelte, Flutter, SwiftUI, Jetpack Compose, React Native
  • Incremental MCP codegen pipeline — codegen_plan, codegen_submit_chunk, codegen_assemble, codegen_clean

Figma Import

  • Import .fig files with layout, fills, strokes, effects, text, images, and vectors preserved

Desktop App

  • Native macOS, Windows, and Linux — a single self-contained binary (winit + GPU Skia, no Electron)
  • .op file association — double-click to open, single-instance lock
  • Background update check against GitHub Releases
  • Native application menu with Save As, Open Recent, and an unsaved-changes dialog on close
  • Recent files persistence

Install

Building from source on Windows: see BUILD_WINDOWS.md

macOS (Homebrew):

brew tap zseven-w/openpencil
brew install --cask openpencil

Windows (Scoop):

scoop bucket add openpencil https://github.com/zseven-w/scoop-openpencil
scoop install openpencil

Linux / Windows direct download: GitHub Releases.exe (Windows), .AppImage / .deb (Linux)

The Windows .exe installer, both Scoop packages, and the install-op.ps1 CLI installer check and, when needed, service the pinned, Microsoft-signed Visual C++ v14 runtime during every install or upgrade. The portable Windows .zip archives and Windows VSIX packages do not run an elevated system installer; install Microsoft's latest supported Visual C++ Redistributable before using those artifacts directly.

Antivirus flagged the Windows installer? That's a known heuristic false positive on new, low-prevalence releases — see how to verify your download.

Nix (Linux x86_64):

nix develop
nix run .                         # launch the desktop app
nix build .#openpencil            # native web host + CanvasKit web bundle
nix build .#op-cli                # the `op` CLI
nix build .#prebuilt              # use the matching upstream desktop archive
nix build .#prebuilt-cli          # use the matching upstream CLI archive
nix build .#web-server            # native GL-free web server + web bundle
nix build .#runtime-prebuilt      # prebuilt desktop + `op` CLI runtime
nix build .#web-sdk-packages      # npm tarballs for the web SDKs
nix build .#appimage              # portable desktop AppImage

The flake uses the pinned Rust toolchain from rust-toolchain.toml and is currently published for x86_64-linux. A Debian package is not produced by the flake yet; use the upstream release artifacts when a .deb is required. The prebuilt outputs use the release version and hashes pinned in nix/release-manifest.json, independently of the workspace source version. After a release is published, the release workflow opens a PR to refresh that manifest. Until it is merged, the prebuilt outputs continue to use the prior published release; the source-built outputs always build the checked-out source.

CLI (op):

brew install zseven-w/openpencil/op

Or use the install script (macOS / Linux):

curl -fsSL https://raw.githubusercontent.com/ZSeven-W/openpencil/main/scripts/install-op.sh | bash

To allow the newest pre-release:

curl -fsSL https://raw.githubusercontent.com/ZSeven-W/openpencil/main/scripts/install-op.sh | OP_PRERELEASE=1 bash

Windows PowerShell:

irm https://raw.githubusercontent.com/ZSeven-W/openpencil/main/scripts/install-op.ps1 | iex

To allow the newest pre-release:

$env:OP_PRERELEASE = "1"; irm https://raw.githubusercontent.com/ZSeven-W/openpencil/main/scripts/install-op.ps1 | iex

Web dev server (builds the CanvasKit wasm bundle, then runs the headless web host)

bash scripts/start-web-rust.sh


Browser-entered built-in provider, image-generation, and Openverse credentials
are kept in same-origin `localStorage`
(`openpencil-rust-web-credentials`) by default,
separate from ordinary web preferences. A credential needed by an AI request is
sent with that request, but the daemon does not write the browser snapshot to
`settings.json`.

Private deployments can opt into server persistence with
`OPENPENCIL_PERSIST_WEB_CREDENTIALS_SERVER=true`. This setting is
deployment-wide; each accepted browser snapshot replaces the shared
browser-owned copy. Enable it only for a trusted, access-controlled deployment
with one active administrator profile, behind HTTPS. Turning the setting off
removes previously persisted browser-owned credentials at the next startup
while preserving operator-owned configuration. As with all `localStorage`,
same-origin scripts can read these values, so the web origin must be protected
from untrusted scripts and XSS. Before downgrading to a release that predates
this switch, turn it off and start the current release once to scrub the
browser-owned copy.

The web host supports built-in API-key agents only. CLI and ACP agents remain
native-only: they are not shown in the web model picker, stored in browser
credential snapshots, synchronized to the server, or exposed through web
connect routes. Browser requests to credential and AI endpoints allow
loopback/localhost origins by default. A public reverse proxy must separately set
`OPENPENCIL_WEB_ALLOWED_ORIGINS` to a comma-separated list of exact origins,
for example `https://demo.openpencil.dev`.

Server persistence does not permit browser-controlled requests to loopback,
private, link-local, metadata, or reserved provider endpoints. A trusted private
deployment that intentionally uses an internal provider must separately set
`OPENPENCIL_WEB_AI_ENDPOINT_ALLOWLIST` to a comma-separated list of exact URL
origins, for example `http://127.0.0.1:11434,https://llm.internal.example`.

Or run as a desktop app:
bash cargo run -p op-host-desktop ```

Prerequisites: Rust (stable) to build the product. Bun >= 1.0 and Node.js >= 18 are only needed for the web SDK under packages/.

Docker

Tagged Rust releases publish a single web-host image. The retired TypeScript images with bundled AI CLIs are no longer published.

ImageIncludes
ghcr.io/zseven-w/openpencil-web:vX.Y.ZRust web host, wasm bundle, and CanvasKit assets

The web UI exposes built-in agent profiles only; Claude/Codex/OpenCode/Copilot CLI tooling is not bundled into Docker images.

Run:

VERSION="$(scripts/workspace-version.sh)"
docker run -d -p 3100:3100 "ghcr.io/zseven-w/openpencil-web:v${VERSION}"

Then open http://localhost:3100/.

Build locally:

docker build -f Dockerfile.web-rust -t openpencil-web-rust .
docker run -p 3100:3100 openpencil-web-rust

Install Rust toolchain (rust-toolchain.toml auto-pins 1.94)

rustup toolchain install 1.94 rustup target add wasm32-unknown-unknown

build / test / lint

cargo check --workspace cargo test --workspace cargo clippy --workspace --all-targets -- -D warnings cargo check --target wasm32-unknown-unknown -p op-collab --no-default-features cargo check --target wasm32-unknown-unknown -p op-host-web --no-default-features --features web # kickoff §1.2 wasm32 invariant cargo deny check && cargo deny --target wasm32-unknown-unknown check bans # cargo-deny (native + wasm32 bans; CI uses cargo-deny-action@v2) ```

Entry crates (crates/ — build targets):

CrateCategorywasm32
op-host-webBrowser bundle entry — wasm32 cdylib, renders via CanvasKit
op-host-nativeNative host lib — WidgetHostNative + skia-safe GL backend (desktop + mobile)❌ (native only)
op-engine-ffiPlayer C ABI — one engine, three shells (iOS Metal / Android EGL / host raster)❌ (native only)
op-engine-jniAndroid JNI marshalling layer for op-engine-ffi (engine thread, registry, callbacks)❌ (native only)
op-host-desktopDesktop binary openpencil-desktop (winit + skia-safe GL) — also the --serve-web daemon that hosts the web bundle❌ (native only)
op-cliop command-line tool❌ (native only)

Shared library crates: op-editor-core (canonical .op state), op-editor-ui (platform-free widgets + RenderBackend), op-editor-host-core (transport-free host state machines), op-collab (open, transport-free collaboration protocol/hash/exact apply), op-collab-transport (native Noise/TCP, bounded admission, and optional LAN discovery), op-mcp, op-ai, op-ai-skills, op-codegen, op-orchestrator, op-figma, op-git, op-opmerge, op-pen-loader, op-design-lint, op-config-store, op-process-io, op-acp, op-i18n, op-rpc-transport — plus op-collab-smoke (authenticated two-process collaboration/recovery matrix) and op-smoke (headless design-turn test runner). The mobile players live in packaging/ (ios + android shells over op-engine-ffi / op-engine-jni). The browser bundle renders through the official CanvasKit skia WASM (loaded separately), so the retired skia-safe-op wasm fork + wasm-libc-shim no longer exist.

Submodule: vendor/agentgithub.com/ZSeven-W/agent-rs (cross-product Rust agent runtime).

🎨 Style Guides

Built-in style guide library with tag-based fuzzy matching. Apply visual styles (glassmorphism, brutalist, retro, etc.) to AI-generated designs. MCP tools for external agent access.

</td> <td width="50%">

Quick Start (Development)

```bash

[OpenPencil × Infistar.cc 无限星河|全模型 API · 驱动 AI 原生设计创作](https://www.infistar.cc/register?aff=LLZC3RLG&ref_source=link)

感谢 Infistar.cc 无限星河 赞助并为 OpenPencil 提供模型服务支持!

  • 稳定承载复杂设计任务: 提供企业级高并发通道与多节点冗余,价格低至官方渠道 1 折,稳定支持长任务、流式响应与连续工具调用。
  • 🧠 一个 API Key 接入主流模型: 全面支持 ChatGPT、Claude、Gemini、Kimi、GLM、DeepSeek 等模型,兼容 OpenAI、Anthropic 等主流接口,可在 OpenPencil 中灵活配置模型服务。
  • 🎨 赋能 AI 原生矢量设计: 适用于提示词生成设计、画布内容修改、多模型协同、图片生成及设计方案迭代,让 AI Agent 更高效地完成从创意到成稿的全过程。
  • 🎁 OpenPencil 用户专属福利: 通过 专属推广链接 注册并完成首次调用,即可领取 5 美元等值测试额度 / 首充专属优惠

⌨️ CLI — `op`

Control the design tool from your terminal. op design, op insert — batch design DSL, node manipulation. Pipe in from files or stdin. Works with desktop app or web server.

</td> </tr> <tr> <td width="50%">

🧩 Embeddable SDK

op-web-sdk (vanilla) + op-web-sdk-react / op-web-sdk-vue adapters — embed a read-only .op viewer in your own app, powered by the same wasm renderer the editor ships.

</td> </tr> <tr> <td width="50%">

CLI — `op`

Install via Homebrew or the install script, then control the design tool from your terminal:

brew install zseven-w/openpencil/op
curl -fsSL https://raw.githubusercontent.com/ZSeven-W/openpencil/main/scripts/install-op.sh | bash
curl -fsSL https://raw.githubusercontent.com/ZSeven-W/openpencil/main/scripts/install-op.sh | OP_PRERELEASE=1 bash
irm https://raw.githubusercontent.com/ZSeven-W/openpencil/main/scripts/install-op.ps1 | iex
$env:OP_PRERELEASE = "1"; irm https://raw.githubusercontent.com/ZSeven-W/openpencil/main/scripts/install-op.ps1 | iex
op start                     # Launch desktop app
op start --headless --file design.op # Launch headless file-backed server
op design @landing.txt       # Batch design from file
op design @ui.js             # Sandboxed JavaScript with loops
op insert '{"type":"rectangle"}' # Insert a node
op import:figma design.fig   # Import Figma file
cat design.dsl | op design - # Pipe from stdin

Supports three input methods: inline string, @filepath (read from file), or - (read from stdin). Works with the desktop app, web server, or headless file-backed server. See the CLI command reference for all commands.

LLM Skill — install the OpenPencil Skill plugin to teach AI agents how to design with op. Run op install for detected agents, or target one explicitly with op install --target codex.

Web SDK / JS tooling (run from packages/)

cd packages && bun run lint # Lint the web SDK (oxlint); also: bun run format cd packages && bun run generate-iconify-catalog # Regenerate the Rust icon catalog assets

Cloning (with submodules)

```bash git clone --recurse-submodules https://github.com/ZSeven-W/openpencil.git

Or already cloned — sync first so stale submodule URLs pick up .gitmodules changes:

git submodule sync --recursive && git submodule update --init --recursive ```

Three submodules live under vendor/, all public and fetched over HTTPS (no SSH key needed): jian (GPU-Skia UI framework — widgets/render/events), casement (winit fork), and agent (agent-rs — cross-product Rust agent runtime, shared by OP + Zode). vendor/anthropic-agent-sdk is tracked in-tree, not a submodule.

🎯 aiskill88 AI 点评 A 级 2026-05-17

创新AI原生设计工具,MCP集成架构前沿。社区热度高、更新活跃,AI赋能设计领域典范之作,具高度参考价值。

📚 实用指南(长尾问题)
适合谁
  • 需要让 Claude / Cursor 操作本地工具的 AI 工程师
  • 构建多智能体协作系统的 Agent 开发者
  • 跨境业务、多语言内容运营团队
最佳实践
  • 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC,避免暴露公网
  • 生产部署优先使用 Docker Compose 隔离依赖,并挂载 volume 持久化数据
  • Agent 任务先做 dry-run 验证工具调用链,再开启自主执行
常见错误
  • API key 直接提交到 git 仓库(请用 .env 并加入 .gitignore)
  • MCP 配置路径拼错或权限不足,重启 Claude Desktop 才生效
  • 容器内无法访问宿主机 localhost — 使用 host.docker.internal
部署方案
  • Docker:openpencil 提供官方镜像,docker compose up 一键启动
  • CLI:直接 npm install -g / pip install,命令行调用
  • 云端托管:可放在 Vercel / Railway / Fly.io 等 PaaS 平台

⚡ 核心功能

  • 通过标准 MCP 协议与 Claude、Cursor 等主流 AI 客户端深度集成
  • 提供结构化工具调用接口,显著降低 AI 集成复杂度
  • 支持 Claude Desktop 和 Claude Code 无缝接入,开箱即用
  • 可与其他 MCP 工具组合叠加,构建完整 AI 工作站
  • 轻量无侵入设计,不影响现有系统架构
👥 适合谁
  • 需要让 Claude / Cursor 操作本地工具的 AI 工程师
  • 构建多智能体协作系统的 Agent 开发者
  • 跨境业务、多语言内容运营团队
⭐ 最佳实践
  • 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC,避免暴露公网
  • 生产部署优先使用 Docker Compose 隔离依赖,并挂载 volume 持久化数据
  • Agent 任务先做 dry-run 验证工具调用链,再开启自主执行
⚠️ 常见错误
  • API key 直接提交到 git 仓库(请用 .env 并加入 .gitignore)
  • MCP 配置路径拼错或权限不足,重启 Claude Desktop 才生效
  • 容器内无法访问宿主机 localhost — 使用 host.docker.internal

👥 适合人群

Claude Desktop / Claude Code 用户AI 工具开发者需要扩展 AI 能力的专业人士自动化工程师

🎯 使用场景

  • 在 Claude Desktop 对话中直接调用本地工具,实现 AI 与系统的深度联动
  • 通过自然语言驱动复杂的多步骤自动化任务,代替繁琐手动操作
  • 将多个 MCP 工具组合使用,构建个人专属 AI 工作站

⚖️ 优点与不足

✅ 优点
  • +MIT 协议,可免费商用
  • +标准化 MCP 协议,生态互联性强
  • +与 Claude 官方生态无缝对接
  • +即插即用,配置简单快捷
⚠️ 不足
  • 依赖 Claude 客户端,非 Claude 用户无法使用
  • MCP 协议仍在持续演进,接口可能变更
  • 需要一定的配置步骤
⚠️ 使用须知

AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。

建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。

📄 License 说明

✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。

❓ 常见问题 FAQ

原生支持Claude及MCP标准协议的AI模型,可通过MCP拓展其他模型能力
💡 AI Skill Hub 点评

总体来看,OpenPencil开源向量设计工具 是一款质量优秀的MCP工具,在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态,建议收藏备用,结合自身场景选择合适时机引入使用。

⬇️ 获取与下载
⬇ 下载源码 ZIP

✅ MIT 协议 · 可免费商用 · 直接从 aiskill88 服务器下载,无需跳转 GitHub

📚 深入学习 OpenPencil开源向量设计工具
查看分步骤安装教程和完整使用指南,快速上手这款工具
🌐 原始信息
原始名称 openpencil
原始描述 开源MCP工具:The world's first open-source AI-native vector design tool and the first to feat。⭐3.1k · TypeScript
Topics 向量设计AI原生MCP协议Agent协作开源设计工具
GitHub https://github.com/ZSeven-W/openpencil
License MIT
语言 TypeScript
🔗 原始来源
🐙 GitHub 仓库  https://github.com/ZSeven-W/openpencil 🌐 官方网站  https://op.zseven.tech

收录时间:2026-05-16 · 更新时间:2026-05-19 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。