Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@agent-incubator-MCP Show me the Agent Incubator dashboard KPIs"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
为 Joule Work Desktop 提供 F&S Agent Incubator 数据查询能力的 MCP（Model Context Protocol）服务器。用户在 Joule 里用自然语言提问，MCP 自动调用 Incubator API 返回结果。
架构： Joule Work → Incubator MCP → Agent Incubator API（afc-srv）
工具 | 说明 |
| 按阶段、业务领域、关键词、负责人、客户搜索 idea，支持分页 |
| 查询单个 idea 的完整信息，包含问卷答案和客户推广记录 |
| 查询总览 KPI：idea 总数、上线数量、各阶段分布 |
Related MCP server: Salesforce MCP Server (Extended)
Node.js 22+
本地 Agent Incubator 实例已启动（默认端口 4004）
npm install
cp .env.example .env编辑 .env：
AUTH_MODE=mock
INCUBATOR_BASE_URL=http://localhost:4004
INCUBATOR_BASIC_AUTH=alice:npm run dev # 在 3000 端口启动 MCP 服务器
npm test # 运行单元测试 + 集成测试打开 Joule Work Desktop
Extensions → Connectors → Add Connector
URL 填 http://localhost:3000/mcp，Auth 选 None
在 Joule 对话里提问，例如："帮我查一下 Agent Incubator 里的数据"
# 列出所有工具
Invoke-RestMethod -Method Post -Uri "http://localhost:3000/mcp" `
-Headers @{"Authorization"="Bearer mock"; "Accept"="application/json, text/event-stream"} `
-ContentType "application/json" `
-Body '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'
# 查询总览数据
Invoke-RestMethod -Method Post -Uri "http://localhost:3000/mcp" `
-Headers @{"Authorization"="Bearer mock"; "Accept"="application/json, text/event-stream"} `
-ContentType "application/json" `
-Body '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_dashboard","arguments":{}}}'
# 搜索 Live 阶段的 idea
Invoke-RestMethod -Method Post -Uri "http://localhost:3000/mcp" `
-Headers @{"Authorization"="Bearer mock"; "Accept"="application/json, text/event-stream"} `
-ContentType "application/json" `
-Body '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"search_ideas","arguments":{"stage":"Live","limit":5}}}'已安装 CF CLI 并登录
cf login -a https://api.cf.eu12.hana.ondemand.com --origin alc4kxudf-platform目标 space：coe-gc-shared-hanacloud / DEV
npm run build
cf push部署完成后地址：
https://agent-incubator-mcp.cfapps.eu12.hana.ondemand.com验证是否正常：
curl https://agent-incubator-mcp.cfapps.eu12.hana.ondemand.com/health
# {"status":"ok","service":"incubator-mcp"}manifest.yml 使用 AUTH_MODE=passthrough——MCP 直接将 Joule 用户的 XSUAA token 转发给 afc-srv，无需单独绑定 XSUAA 服务实例（前提：MCP 和 Incubator 在同一个 BTP subaccount）。
部署完成后，在 BTP Cockpit → Connectivity → Destinations 新建：
字段 | 值 |
Name |
|
URL |
|
Authentication |
|
Additional Property |
|
然后在 Joule Work Desktop：
Extensions → Connectors → Add Connector
选择 BTP Destination agent-incubator-mcp
变量 | 默认值 | 说明 |
|
| 监听端口 |
|
|
|
|
| Agent Incubator API 地址 |
| — | 本地 CAP 开发服务器的 Basic Auth，格式 |
| — | 仅 |
| — | 仅 |
| — | 仅 |
模式 | 适用场景 |
| 本地开发，不需要真实认证 |
| CF 部署，直接转发 Joule 的 XSUAA token（同一 BTP subaccount） |
| 跨 subaccount，需要用 XSUAA 凭据做 JWT bearer exchange |
This server cannot be installed
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Capture, analyze, validate, and submit inventions, and collaborate on patent reviews.
An AI concierge that turns static forms into adaptive AI conversations. From any MCP client.
Find relevant security data from Sentinel data lake for building effective agents. More:aka.ms/s/de
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/xinlyu02/agent-incubator-MCP'
If you have feedback or need assistance with the MCP directory API, please join our Discord server