@belal-elsabbagh-apex/copilot-mcp
Enables EHR Copilot operations such as order cloning, queue building, and execution trace analysis over stdio.
README
@belal-elsabbagh-apex/copilot-mcp
MCP server exposing EHR Copilot operations over stdio. Built with Bun
(bun build bundles src/ → a node-runnable dist/server.js), so it runs under either
bunx or npx. Published as a private package on GitHub Packages.
Tools
| Tool | What it does |
|---|---|
clone_order |
Mirror PROD order(s) into PRE-PROD as fresh orders (clone-only by default; submit is opt-in). |
find_clone_candidates |
List recent PROD orders that will actually clone to forReview. |
delete_preprod_order |
Delete order(s) from PRE-PROD (never touches prod). |
build_queue_item |
Build a UiPath AddQueueItem request (payload + curl) from an order. BUILD ONLY — IsApproved is always false. |
analyze_order_execution |
Trace an order to its UiPath Orchestrator job(s) and diagnose the run. READ-ONLY. |
Configuration
The server reads one validated config holding both the Copilot BE creds and the UiPath args. Two ways to provide it:
- Single file (preferred): set
COPILOT_MCP_CONFIGto a JSON file shaped likeconfig.example.json. - Split legacy files: set
COPILOT_MCP_LOCAL_DIRto a directory containingorder-copy-credentials.json+uipath-config.json(+ optionaloverrides.json).
The uipath.noteBucket / queueUrl / addQueueItemPath / serverUrlByEnv fields are only
required by build_queue_item. overrides (per-prodUid clone remaps) is optional.
Other env vars: COPILOT_MCP_DEBUG_DIR (where clone_order dumps the extracted prod
order JSON; defaults to the OS temp dir).
Local development
bun install
bun run typecheck
bun run build # bundle src -> dist/server.js (node-runnable, what npx/bunx execute)
bun run start # run from source via bun (no build needed)
bun run dev # watch mode
Installing from GitHub Packages (private)
The package is private, so installing requires a GitHub token with read:packages.
Add to the consumer repo's .npmrc:
@belal-elsabbagh-apex:registry=https://npm.pkg.github.com
//npm.pkg.github.com/:_authToken=${NODE_AUTH_TOKEN}
Then export NODE_AUTH_TOKEN=<your PAT> and install:
bun add -d @belal-elsabbagh-apex/copilot-mcp
Wire into .mcp.json
{
"mcpServers": {
"copilot": {
"command": "npx",
"args": ["-y", "@belal-elsabbagh-apex/copilot-mcp"],
"env": {
"COPILOT_MCP_CONFIG": "/abs/path/to/config.local.json"
}
}
}
}
npx and bunx are interchangeable here (the published bin, dist/server.js, is a
plain node ESM entry). Use COPILOT_MCP_LOCAL_DIR instead of COPILOT_MCP_CONFIG for
the split legacy files.
Publishing
Publishing is tag-driven. Bump version in package.json, commit, then push a matching
vX.Y.Z tag:
git tag v1.2.0
git push origin v1.2.0
That one tag push fans out to two workflows (both keyed off the tag, because a
GITHUB_TOKEN-created release does not trigger other workflows):
- Release on tag (
release.yml) — creates a GitHub Release with auto-generated notes. - Publish (
publish.yml) — typechecks, builds, and runsbun publish --access restrictedusing the repo'sGITHUB_TOKEN(packages: write), keeping the package private. Also runnable manually via workflow_dispatch.
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。