Shortcut MCP Server
Provides comprehensive tools for interacting with the Shortcut API, enabling developers to manage stories, comments, projects, and analyze Loom videos for debugging insights.
README
Shortcut MCP Server
A Model Context Protocol (MCP) server that provides comprehensive tools for interacting with the Shortcut API and analyzing Loom videos attached to stories. This server helps developers:
- Read and update stories, manage comments, and query project data
- Analyze Loom videos for debugging insights
- Correlate video content with codebase to identify relevant files
- Generate debugging plans based on video analysis
Features
- Story Management: Get, search, and update stories
- Comment System: Add and retrieve story comments
- Project Organization: List projects, epics, and workflows
- File Attachments: Download and analyze file attachments from stories
- Video Analysis: Extract and analyze Loom videos from stories for debugging
- Codebase Correlation: Match video content with relevant source files
- Debugging Assistance: Generate debugging plans from video analysis
- Type Safety: Full TypeScript support with comprehensive type definitions
- Error Handling: Robust error handling with detailed error messages
- Response Size Management: Automatic warnings for large responses (>20k tokens)
Installation
- Clone the repository:
git clone <repository-url>
cd shortcut-mcp
- Install dependencies using pnpm:
pnpm install
- Create the configuration directory and file:
mkdir -p ~/.shortcut_mcp
echo "SHORTCUT_TOKEN=your-api-token-here" > ~/.shortcut_mcp/.env
- Build the project:
pnpm run build
Configuration
The server reads the Shortcut API token from ~/.shortcut_mcp/.env. This file should contain:
SHORTCUT_TOKEN=your-shortcut-api-token
To obtain your Shortcut API token:
- Log in to Shortcut
- Navigate to Settings > API Tokens
- Generate a new token
- Copy the token to your .env file
Usage
Start the MCP server:
./start_mcp.sh
For development mode with hot reload:
./start_mcp.sh dev
For hot reload support that maintains MCP connection:
./start_mcp_with_proxy.sh
# In another terminal, trigger reload with:
./reload_mcp.sh
Available Tools
Story Operations
shortcut_get_story- Get detailed information about a specific storyshortcut_search_stories- Search for stories using Shortcut's query syntaxshortcut_get_project_stories- Get all stories in a specific projectshortcut_get_epic_stories- Get all stories in a specific epicshortcut_update_story- Update story attributes (description, labels, workflow state, etc.)
Comment Operations
shortcut_add_comment- Add a comment to a storyshortcut_get_comments- Retrieve all comments for a story
Organization Tools
shortcut_list_workflows- List all workflows and their statesshortcut_list_projects- List all projects in the workspaceshortcut_list_epics- List all epics in the workspace
File Operations
shortcut_get_story_files- Get all file attachments for a storyshortcut_download_file- Download file content from attachments (text or binary)
Video Analysis Tools
shortcut_get_story_loom_videos- Extract Loom video links from a storyshortcut_analyze_loom_video- Analyze Loom video metadata, transcript, and debugging insightsshortcut_analyze_video_with_codebase- Correlate video content with codebase filesshortcut_debug_video_issue- Deep debugging analysis with action tracking and error detectionshortcut_analyze_video_with_anthropic- (Currently disabled) Visual analysis using Claude Vision
Development
Scripts
pnpm run dev- Run in development mode with hot reloadpnpm run build- Build TypeScript to JavaScriptpnpm run typecheck- Check TypeScript typespnpm run lint- Run ESLintpnpm run format- Format code with Prettierpnpm test- Run unit and integration tests
Project Structure
shortcut-mcp/
├── src/
│ ├── server.ts # Main MCP server implementation
│ ├── shortcutClient.ts # Shortcut API client
│ ├── shortcut-types.ts # TypeScript type definitions
│ ├── loomClient.ts # Loom API integration
│ ├── loomDetector.ts # Loom URL detection utilities
│ ├── videoAnalysis.ts # Core video analysis service
│ ├── videoCodeAnalyzer.ts # Codebase correlation analysis
│ ├── enhancedVideoAnalyzer.ts # Deep debugging analysis
│ ├── audioAnalyzer.ts # Audio transcript analysis
│ ├── visualAnalyzer.ts # Visual element detection
│ ├── debuggingAssistant.ts # Debugging plan generation
│ ├── proxy.ts # Hot reload proxy server
│ └── mimeTypes.ts # File type utilities
├── tests/ # Test files
│ ├── unit/ # Unit tests
│ └── integration/ # Integration tests
├── dist/ # Compiled JavaScript (generated)
├── start_mcp.sh # Standard server startup script
├── start_mcp_with_proxy.sh # Hot reload startup script
├── reload_mcp.sh # Trigger hot reload
├── package.json # Project configuration
└── tsconfig.json # TypeScript configuration
Error Handling
The server provides detailed error messages for common issues:
- Missing or invalid API token
- API request failures
- Invalid story/project/epic IDs
- Network connectivity issues
License
ISC
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。