serve7-mcp-connectors
Serve7 MCP Connector Library
为 Amazon Connect AI Agents 和 Bedrock AgentCore 提供的可复用 MCP 工具。七个通用连接器——客户查询、创建支持工单、检查工单状态、安排预约、更新 CRM 记录、检索知识及创建任务——在这里一次性构建,然后在 Serve7 的每个客户交互中复用,而无需在每个项目中从头接入。
这是 2026 年秋季项目点子列表中的项目 #6(“Serve7 MCP Connector Library”)。它是一个可运行的第一版:目前每个连接器都能在内存模拟后端上端到端运行(无需 AWS 账户、无需真实 CRM、无需凭据即可试用),并留有清晰的接缝,便于日后接入真实后端。
为什么用适配器,而不是直接集成
每个连接器的业务逻辑(结构校验、什么算错误,以及输入/输出字段)都位于 src/connectors/*.ts 中,并且不会直接调用具体后端的 SDK。它调用的是 src/adapters/types.ts 中五个小接口——CrmAdapter、TicketingAdapter、SchedulingAdapter、KnowledgeAdapter、TaskAdapter。目前它们全部在内存中实现(src/adapters/mock.ts)。日后要接入真实的 Salesforce/ServiceNow/Connect-Cases/Bedrock-Knowledge-Base 等后端,只需编写一个新的适配器类并改动一处 switch 语句(src/adapters/index.ts)——连接器代码、其测试,或智能体见到的 MCP 工具定义均不必改动。参见 docs/adding-a-real-adapter.md。
MCP client (Claude Desktop, Bedrock AgentCore, ...)
│ tool call: customer_lookup({ customerId: "cust-1001" })
▼
src/server.ts (stdio) or src/http-server.ts (deployed)
│ registers every connector as an MCP tool
▼
src/connectors/customer-lookup.ts
│ validates input (zod), calls adapters.crm.lookupCustomer(...)
▼
src/adapters/index.ts → mock.ts today, a real adapter laterRelated MCP server: Runtime MCP Server
项目结构
src/
connectors/ one file per MCP tool — schema + business logic only
adapters/ the CRM/ticketing/scheduling/knowledge/task interfaces,
plus the in-memory mock implementations
mock-data/ seed data the mock adapters serve
mcp-server-factory.ts registers every connector on an McpServer instance
server.ts stdio entrypoint (local dev, Claude Desktop, Claude Code)
http-server.ts Streamable-HTTP entrypoint (deployed, e.g. behind Lambda)
tests/ one test file per connector + a registry sanity test
deploy/ Dockerfile + AWS SAM template for a Lambda deployment
docs/ per-connector reference + how to add a real adapter在本地运行
npm install
npm test # 29 tests, all against the mock adapters
npm run typecheck # strict TypeScript, zero `any` in connector code
npm start # stdio MCP server — Ctrl+C to stop
npm run start:http # HTTP MCP server on :8080 (curl http://localhost:8080/health)
npm run smoke # real MCP client over stdio, calls all 7 tools end to end
npm run smoke:http # same, over HTTP — start:http must be running first接入 Claude Desktop / Claude Code
将 MCP 客户端指向 stdio 服务器即可,例如,在 Claude Desktop 的 claude_desktop_config.json 中配置:
{
"mcpServers": {
"serve7-connectors": {
"command": "npx",
"args": ["tsx", "/absolute/path/to/serve7-mcp-connectors/src/server.ts"]
}
}
}然后可以提问,比如“查询客户 cust-1001”或“我们有哪些关于紧急停业通知的知识文章?”——它会在模拟数据上调用 customer_lookup / retrieve_knowledge。
为真实场景部署(例如,通过 Bedrock AgentCore)
参见 deploy/README.md —— 通过 sam deploy 构建一个容器镜像 Lambda,并将其置于一个 IAM 认证的 Function URL 之后。
添加第 8 个连接器
使用
defineConnector()创建src/connectors/my-new-tool.ts,可以复制一个现有连接器的形态,例如create-task.ts。将其添加到
src/connectors/index.ts中的CONNECTORS数组。在
tests/中添加测试文件。在
docs/connectors.md中为其添加文档。
server.ts、http-server.ts 以及 registry 测试都会自动发现它,此外无需再改任何文件。
本第一版刻意不包括的内容
根据原始项目说明,每个连接器最终还应包括认证方案(已完成,位于适配器层)、部署模板(已完成,deploy/)和文档(已完成,docs/)——这些已全部交付。尚未实现、值得快速跟进的是:至少为一种后端提供真实适配器(这样它就不全是模拟实现),以及将可运行的 Bedrock AgentCore 智能体通过已部署的 HTTP 端点到端地接入。
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- FlicenseNot gradedqualityBmaintenanceEnables interaction with Salesforce data and services via custom MCP tools, including account analytics, opportunity queries, case creation, and AI agent invocation.4
- FlicenseNot gradedqualityCmaintenanceDeploys and manages MCP tools on Amazon Bedrock AgentCore Runtime, enabling knowledge base retrieval and AWS infrastructure management through Streamlit applications.
- AlicenseAqualityFmaintenanceProvides 14 MCP tools for AI agent infrastructure, enabling knowledge base queries, skill search, handoffs, blueprint validation, trust scoring, identity verification, SLA validation, and compliance checks.22MIT
- FlicenseNot gradedqualityCmaintenanceProvides Claude agents with CRM contact lookup, action logging, and prompt cost auditing tools. Includes a Streamlit UI to demo the same tools without an MCP client.
Related MCP Connectors
LeadConnector / GoHighLevel MCP Pack — wraps the GoHighLevel CRM for AI agents.
Operator-as-agent MCP hub. 6 tools. First $5 free, then $0.001/call.
Search, document and execute authenticated API calls across 700+ apps via one MCP server
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/MilosJova/serve7-mcp-connectors'
If you have feedback or need assistance with the MCP directory API, please join our Discord server