TypeScript로 작성된 최소 구성의 Model Context Protocol (MCP) 서버입니다. 소스에서 직접 빌드하고 Docker로 컨테이너화하여 IBM Code Engine에 배포할 수 있도록 설계되었습니다.
Related MCP server: My Awesome MCP
이 서버는 두 가지 transport를 지원합니다:
Transport | 사용되는 시점 | 활성화 방법 |
stdio | 로컬 개발 / Bob 데스크톱 | 기본값(환경 변수 불필요) |
HTTP Streamable | Docker / Code Engine |
|
HTTP transport는 liveness probe를 위해 GET /health에서 /health 엔드포인트를 노출합니다.
.
├── src/
│ └── index.ts # Server entry point — add tools here
├── build/ # Compiled output (git-ignored)
├── .github/
│ └── workflows/
│ └── build-push.yml # CI: typecheck → build → push to GHCR
├── Dockerfile # Multi-stage build (builder + runtime)
├── docker-compose.yml # Local container testing
├── tsconfig.json
└── package.json도구 | 설명 |
| 상태 확인 — |
| 서버 이름, 버전, transport, Node 버전, 업타임(uptime)을 반환합니다 |
Node.js 20+
npm 10+
npm install
npm run buildnode build/index.jsMCP_TRANSPORT=http PORT=3000 node build/index.js헬스 엔드포인트를 테스트합니다:
curl http://localhost:3000/healthdocker build -t erm-github-mcp .docker run -p 3000:3000 erm-github-mcpdocker compose up --build.github/workflows/build-push.yml 워크플로우는 main 브랜치에 대한 모든 push와 버전 태그(v*)에서 실행됩니다:
TypeScript 빌드 — npm ci + tsc (타입 오류 시 즉시 실패)
Docker build & push — 멀티 플랫폼 이미지(linux/amd64, linux/arm64)를 GitHub Container Registry (GHCR)에 push합니다
이미지는 다음과 같이 게시됩니다:
ghcr.io/<your-github-username>/erm-github-mcp:<tag>구성해야 할 secrets는 없습니다. 워크플로우는 내장된 GITHUB_TOKEN을 사용합니다.
첫 번째 push 후 GitHub → Packages → erm-github-mcp → Package settings로 이동하여 표시 여부를 Public으로 설정하세요. 그래야 Code Engine이 자격 증명 없이 pull할 수 있습니다.
IBM Cloud CLI와 Code Engine 플러그인:
ibmcloud plugin install code-engineIBM Cloud 계정과 Code Engine 프로젝트
ibmcloud ce project select --name <your-project-name>배포하려는 이미지 태그(예: main 또는 v0.1.0)로 <tag>를 대체하세요:
ibmcloud ce application create \
--name erm-github-mcp \
--image ghcr.io/<your-github-username>/erm-github-mcp:<tag> \
--port 3000 \
--min-scale 0 \
--max-scale 5 \
--env MCP_TRANSPORT=http \
--env PORT=3000ibmcloud ce application get --name erm-github-mcp --output urlibmcloud ce application update \
--name erm-github-mcp \
--image ghcr.io/<your-github-username>/erm-github-mcp:<new-tag>변수 | 기본값 | 설명 |
|
| 컨테이너에서는 |
|
| 서버가 수신하는 HTTP 포트 |
Bob의 mcp.json에 다음을 추가하세요:
{
"mcpServers": {
"erm-github-mcp": {
"url": "https://<your-code-engine-url>"
}
}
}{
"mcpServers": {
"erm-github-mcp": {
"command": "node",
"args": ["/absolute/path/to/erm-github-mcp/build/index.js"]
}
}
}src/index.ts에서 server.tool()을 사용하여 새 도구를 추가하세요:
server.tool(
"my_tool",
"Description shown to the AI",
{
input_field: z.string().describe("What this field does"),
},
async ({ input_field }) => ({
content: [{ type: "text", text: `Result: ${input_field}` }],
})
);그런 다음 다시 빌드하세요:
npm run buildReturns metadata about this running MCP server instance.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. 'Returns metadata' implies a read-only operation, but it does not explicitly say so, nor does it disclose what the response contains or whether there are side effects. For a zero-parameter tool this is adequate, but it misses the chance to confirm non-mutating behavior and response shape.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single, front-loaded sentence with no wasted words. It states the verb, resource, and scope efficiently. Perfect for a zero-parameter tool.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple read-only metadata tool with no parameters and no output schema, the description is mostly complete. It tells the agent what the tool does. However, it could be slightly more explicit about the types of metadata returned (e.g., server version, protocol, capabilities) and could mention that it is safe to call without side effects. The lack of any relationship to 'ping' also leaves a small gap.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has zero parameters, so there is nothing for the description to explain beyond what the schema already shows (empty properties). The baseline is 4 because the schema fully covers parameters, and the description adds no ambiguity. No extra value needed.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a clear verb ('Returns') and resource ('metadata about this running MCP server instance'). It distinguishes from ping (which implies liveness check), though 'metadata' is generic and could be more specific about what metadata (e.g., version, capabilities). Still, the purpose is unmistakable.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance on when to use this vs the sibling tool 'ping'. Both relate to server status, but the description doesn't explain which to call for what. An agent would have to infer that ping checks connectivity while this fetches metadata, but no explicit when/when-not is provided.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Health-check tool. Returns 'pong' and echoes the optional message back.
| Name | Required | Description | Default |
|---|---|---|---|
| message | No | Optional message to echo back |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden, and it discloses both the return value ('pong') and the side effect (echoing the message). 'Health-check' strongly implies a non-mutating, safe operation. It does not explicitly state non-destructiveness, but for a ping the disclosed behavior is effectively complete.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two short sentences, no filler, with the purpose front-loaded before the return-behavior detail. Every word earns its place; this is a model of concise health-check documentation.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a tool this simple — one optional parameter, no required fields, no nested objects, no output schema — the description tells an agent everything needed to call it correctly and predict the response. Nothing meaningful is missing.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the single message parameter is already documented ('Optional message to echo back'). The description reinforces the echo behavior in prose, which adds marginal context beyond the schema but does not materially expand on it, landing at the baseline for high-coverage schemas.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific purpose ('health-check tool') and the precise behavior ('returns pong and echoes the optional message back'). This is clear, specific, and inherently distinguishes it from the get_server_info sibling, whose name implies returning server details rather than a liveness signal.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The 'health-check tool' phrasing implies when it should be used, and it is self-evident for a ping utility. However, there is no explicit guidance on when to prefer this over get_server_info or any stated exclusions. For a trivial tool the context is clear enough, but it stops short of explicit alternative routing.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0get_server_infopingScored across 2 tools
The two tools serve clearly distinct purposes: ping is a health check, while get_server_info provides instance metadata. There is zero overlap between them.
Both tools follow a consistent verb_noun naming pattern (ping, get_server_info) with snake_case, matching the expected convention.
With only two generic tools and a server name indicating GitHub integration, the count is severely inadequate. The tools are trivial and unrelated to the apparent domain, representing an extreme mismatch.
The server name implies GitHub functionality, but the tool surface offers only health check and server info. No GitHub operations exist, leaving obvious and critical gaps.
An MCP server that let you interact with Cycloid.io Internal Development Portal and Platform
MCP server for AI agents to plan, verify, and deploy Cloudflare-native apps.
An MCP server for Arcjet - the runtime security platform that ships with your AI code.
The official MCP Server from Mia-Platform to interact with Mia-Platform Console
