Squad

效率与工作流

by the-basilisk-ai

你的 AI Product Manager,借助 30 多种工具挖掘洞察、制定路线图并规划产品战略。

什么是 Squad

你的 AI Product Manager,借助 30 多种工具挖掘洞察、制定路线图并规划产品战略。

README

Squad MCP Server

A remote MCP server that brings Squad — the AI-powered product discovery and strategy platform — directly into your AI workflows. Connect Squad to Claude, ChatGPT, or any MCP-compatible AI assistant to research, ideate, and plan products without context switching.

🚀 Quick Start

For Users

Connect Squad to your AI assistant in seconds:

Claude Code:

bash
claude mcp add --transport http squad https://mcp.meetsquad.ai/mcp

On first use, you'll be prompted to authenticate via OAuth in your browser.

Claude Connectors:

  • Coming soon to the Claude MCP directory

ChatGPT:

  • Coming soon to the ChatGPT plugin store

Other MCP Clients:

Connect using https://mcp.meetsquad.ai/mcp - OAuth configuration is automatically discovered via the server's .well-known/oauth-authorization-server endpoint.

📖 Usage Examples

See USAGE_EXAMPLES.md for detailed real-world examples including:

  • Discover opportunities - "What opportunities are in my workspace?"
  • Explore solutions - "Show me solutions for [opportunity] with pros/cons"
  • Strategic alignment - "How do my solutions map to business goals?" (OST view)
  • Generate ideas - "Generate solution ideas for [opportunity]"
  • Search everything - "Find all content related to compliance"
  • Create opportunities - "Create a new opportunity for [customer pain point]"

Each example shows the actual user prompt, which tools get called behind the scenes, and the expected output based on real Squad data.

✨ Available Tools

The Squad MCP server provides 30+ tools across 9 categories:

CategoryToolsPurpose
Opportunitieslist_opportunities, get_opportunity, create_opportunity, update_opportunity, delete_opportunity, generate_solutions, manage_opportunity_relationshipsDiscover and refine product opportunities
Solutionslist_solutions, get_solution, create_solution, update_solution, delete_solution, manage_solution_relationships, prioritise_solutionsGenerate and iterate on solution ideas
Goalslist_goals, get_goal, create_goal, update_goal, delete_goal, manage_goal_relationshipsDefine and track business objectives
Knowledgelist_knowledge, get_knowledge, create_knowledge, delete_knowledgeStore research, references, and insights
Feedbacklist_feedback, get_feedback, create_feedback, delete_feedbackManage customer and stakeholder feedback
Insightslist_insights, get_insight, create_insight, delete_insightTrack customer insights and feature requests
Workspacelist_workspaces, select_workspace, get_workspace, update_workspaceConfigure workspace settings
Searchsimilarity_searchSemantic search across all entities
Viewsview_strategy_context, view_roadmapRich visual strategy and roadmap widgets

Tool Capabilities

All tools include:

  • ✅ Safety annotations (readOnlyHint / destructiveHint)
  • ✅ Structured JSON schemas for inputs/outputs
  • ✅ User-isolated data access via OAuth
  • ✅ Relationship management between entities

🏗️ Architecture

code
┌─────────────┐         OAuth          ┌──────────────┐
│  Claude /   │ ◄────────────────────► │  PropelAuth  │
│  ChatGPT    │    (Authentication)     │   (IdP)      │
└─────────────┘                         └──────────────┘
       │
       │ HTTPS + Bearer Token
       ▼
┌──────────────────────────────────────────────┐
│  Squad MCP Server                            │
│  ┌────────────────────────────────────────┐  │
│  │  OAuth Middleware → Validate Token     │  │
│  │  JWT Minting → Service Credentials    │  │
│  │  Session Store → Manage State          │  │
│  │  MCP Handler → Execute Tools           │  │
│  └────────────────────────────────────────┘  │
└──────────────────────────────────────────────┘
       │
       │ Squad API Calls (minted JWT)
       ▼
┌──────────────┐
│  Squad API   │
└──────────────┘

📦 NPM Package

For programmatic access to Squad tools in your Node.js applications:

bash
npm install @squadai/tools
typescript
import { tools as squadTools } from "@squadai/tools";

// Use with Vercel AI SDK
const result = await generateText({
  model: anthropic("claude-3-5-sonnet-20241022"),
  tools: squadTools({
    jwt: "YOUR_JWT_TOKEN",
    orgId: "org-123",
    workspaceId: "ws-456",
  }),
  prompt: "List my current product opportunities",
});

🛠️ Development

This repository contains the source code for the Squad MCP remote server.

Prerequisites

  • Node.js 22.22+
  • Yarn
  • Nix (optional, for reproducible dev environment via flake.nix)
  • PropelAuth account (for OAuth2)
  • Squad API credentials

Local Setup

bash
# Clone repository
git clone https://github.com/the-basilisk-ai/squad-mcp.git
cd squad-mcp

# Install dependencies
yarn install

# Configure environment
cp .env.example .env
# Edit .env with your PropelAuth credentials

# Start development server with hot reload
yarn dev

# Server available at http://localhost:3232

Available Commands

bash
yarn build              # Compile TypeScript
yarn dev                # Start dev server with hot reload
yarn start              # Start production server
yarn test               # Run unit tests (vitest)
yarn format             # Lint with biome
yarn format:fix         # Auto-fix lint issues
yarn openapi:squad      # Regenerate API client from OpenAPI spec
yarn storybook          # Run Storybook for widget development

Testing the Server

bash
# Check health
curl http://localhost:3232/health

# Check OAuth discovery
curl http://localhost:3232/.well-known/oauth-authorization-server

# Test with MCP Inspector
npx @modelcontextprotocol/inspector
# Then connect to http://localhost:3232/mcp

Project Structure

code
squad-mcp/
├── index.ts                    # MCP server entry point with OAuth
├── server.json                 # MCP registry metadata
├── src/
│   ├── client.ts               # MCP client export
│   ├── helpers/
│   │   ├── config.ts           # Environment configuration
│   │   ├── getUser.ts          # OAuth context + workspace selection
│   │   └── mintToken.ts        # JWT minting + per-user cache
│   ├── lib/
│   │   ├── logger.ts           # Structured logging
│   │   └── clients/
│   │       ├── squad.ts        # Squad API client factory
│   │       └── middleware/     # Bearer token middleware
│   └── tools/                  # Tool implementations
│       ├── opportunity.ts
│       ├── solution.ts
│       ├── goal.ts
│       ├── knowledge.ts
│       ├── feedback.ts
│       ├── insight.ts
│       ├── workspace.ts
│       ├── search.ts
│       ├── views.ts
│       └── helpers.ts
├── resources/                  # React widget components
│   ├── view-strategy-context/
│   └── view-roadmap/
├── railway.toml                # Railway deployment config
└── .env.example                # Environment template

🏭 Production Deployment

This is a hosted service maintained by Squad. Users connect via OAuth - no self-hosting required.

Architecture Notes (for contributors):

  • Single-instance deployment on Railway
  • Follows MCP specification for stateful HTTP sessions
  • In-memory transport storage (standard per MCP spec)

💬 Support

Need help with the Squad MCP server?

🤝 Contributing

Contributions welcome! Pre-commit hooks run biome lint and vitest automatically. Please ensure:

  • yarn format passes (biome lint)
  • yarn build compiles without errors
  • yarn test passes
  • All tools include safety annotations

📄 License

MIT

🔗 Links

常见问题

Squad 是什么?

你的 AI Product Manager,借助 30 多种工具挖掘洞察、制定路线图并规划产品战略。

相关 Skills

表格处理

by anthropics

Universal
热门

围绕 .xlsx、.xlsm、.csv、.tsv 做读写、修复、清洗、格式整理、公式计算与格式转换,适合修改现有表格、生成新报表或把杂乱数据整理成交付级电子表格。

做 Excel/CSV 相关任务很省心,能直接读写、修复、清洗和格式转换,尤其擅长把乱七八糟的表格整理成交付级文件。

效率与工作流
未扫描109.6k

PDF处理

by anthropics

Universal
热门

遇到 PDF 读写、文本表格提取、合并拆分、旋转加水印、表单填写或加解密时直接用它,也能提取图片、生成新 PDF,并把扫描件通过 OCR 变成可搜索文档。

PDF杂活别再来回切工具了,文本表格提取、合并拆分到OCR识别一次搞定,连扫描件也能变可搜索。

效率与工作流
未扫描109.6k

Word文档

by anthropics

Universal
热门

覆盖Word/.docx文档的创建、读取、编辑与重排,适合生成报告、备忘录、信函和模板,也能处理目录、页眉页脚、页码、图片替换、查找替换、修订批注及内容提取整理。

搞定 .docx 的创建、改写与精排版,目录、批量替换、批注修订和图片更新都能自动化,做正式文档尤其省心。

效率与工作流
未扫描109.6k

相关 MCP Server

文件系统

编辑精选

by Anthropic

热门

Filesystem 是 MCP 官方参考服务器,让 LLM 安全读写本地文件系统。

这个服务器解决了让 Claude 直接操作本地文件的痛点,比如自动整理文档或生成代码文件。适合需要自动化文件处理的开发者,但注意它只是参考实现,生产环境需自行加固安全。

效率与工作流
82.9k

by wonderwhy-er

热门

Desktop Commander 是让 AI 直接执行终端命令、管理文件和进程的 MCP 服务器。

这工具解决了 AI 无法直接操作本地环境的痛点,适合需要自动化脚本调试或文件批量处理的开发者。它能让你用自然语言指挥终端,但权限控制需谨慎,毕竟让 AI 执行 rm -rf 可不是闹着玩的。

效率与工作流
5.8k

EdgarTools

编辑精选

by dgunning

热门

EdgarTools 是无需 API 密钥即可解析 SEC EDGAR 财报的开源 Python 库。

这个工具解决了金融数据获取的痛点——直接让 AI 读取结构化财报,比如让 Claude 分析苹果的 10-K 文件。适合量化分析师或金融开发者快速构建数据管道。但注意,它依赖 SEC 网站稳定性,高峰期可能延迟。

效率与工作流
1.9k

评论