feat(image-gen): 移除 header 传 model 的逻辑,仅保留工具参数 → 兜底默认
- 删除 extractModelFromHeaders / X-Image-Gen-Model 等 header 解析路径 - model 优先级简化为单链:arguments.model → DEFAULT_MODEL (CogView-3-Flash) - 新增 README.md,覆盖启动、客户端配置、工具用法、curl 调试、FAQ
This commit is contained in:
@@ -0,0 +1,237 @@
|
||||
# MCP Image Gen
|
||||
|
||||
一个基于 [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) 的图片生成服务,对接智谱 BigModel 的 `CogView` 系列模型。每次调用通过 HTTP header 传入智谱 API 的 Bearer Token,支持在工具参数中按需切换模型。
|
||||
|
||||
- 协议:MCP **Streamable HTTP**(单端点 `/mcp`,支持 SSE 流)
|
||||
- 上游:[智谱 BigModel Images Generations API](https://open.bigmodel.cn/dev/api#images)
|
||||
- 默认模型:`CogView-3-Flash`
|
||||
- 端口:`3000`
|
||||
- 鉴权:**无服务端密钥**,每次请求必须从 header 携带智谱 token
|
||||
|
||||
---
|
||||
|
||||
## 快速开始
|
||||
|
||||
### 启动(docker-compose)
|
||||
|
||||
仓库根目录 `docker-compose.yml` 已包含本服务:
|
||||
|
||||
```bash
|
||||
cd /root/workspace/tools/mcp
|
||||
docker compose up -d --build mcp-image-gen
|
||||
```
|
||||
|
||||
服务容器名 `mcp-image-gen`,通过反代暴露为 `https://image-gen.ai.foresh.com/mcp`。
|
||||
|
||||
直接连容器调试:
|
||||
|
||||
```bash
|
||||
docker exec -it mcp-image-gen sh
|
||||
# 或本机
|
||||
curl http://localhost:3000/health
|
||||
# {"status":"ok","service":"image-gen","sessions":0}
|
||||
```
|
||||
|
||||
### MCP 客户端配置
|
||||
|
||||
#### Claude Desktop / Cursor 等支持 `headers` 配置的客户端
|
||||
|
||||
```jsonc
|
||||
// Claude Desktop: ~/Library/Application Support/Claude/claude_desktop_config.json
|
||||
// Cursor: ~/.cursor/mcp.json
|
||||
{
|
||||
"mcpServers": {
|
||||
"image-gen": {
|
||||
"url": "https://image-gen.ai.foresh.com/mcp",
|
||||
"headers": {
|
||||
"Authorization": "Bearer <your-zhipu-api-key>"
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
> **重要**:客户端必须在 **每一个 HTTP 请求**(`initialize` / `notifications/initialized` / `tools/call`)都带上 `Authorization` header。配置在 `headers` 里、由客户端自动重发即可。直连模式下 token 不在 session 间共享。
|
||||
|
||||
#### 不支持 `headers` 字段的客户端
|
||||
|
||||
如果客户端只支持 `command`/`args` 启动方式,可改用 stdio 模式(需自行包一层把 token 注入环境变量,再让工具回调从环境变量读——当前实现 HTTP 模式下不依赖环境变量,需自行扩展)。
|
||||
|
||||
---
|
||||
|
||||
## 工具:generate_image
|
||||
|
||||
调用智谱图片生成接口,返回图片 URL。
|
||||
|
||||
### 参数
|
||||
|
||||
| 名称 | 类型 | 必填 | 说明 |
|
||||
|---|---|---|---|
|
||||
| `prompt` | string | ✅ | 图片生成提示词,描述想要的画面内容 |
|
||||
| `model` | string | ❌ | 模型名称,默认 `CogView-3-Flash`。可填 `CogView-3` / `CogView-3-Flash` 等智谱支持的模型 |
|
||||
| `size` | string | ❌ | 图片尺寸,例如 `1024x1024`、`1280x1280`、`1024x1792`、`1792x1024`、`768x1344`、`1344x768`,遵循智谱 API 规格 |
|
||||
|
||||
### Model 优先级(单链)
|
||||
|
||||
```
|
||||
工具参数 arguments.model → DEFAULT_MODEL (CogView-3-Flash)
|
||||
```
|
||||
|
||||
每次 `tools/call` 都是独立决策,**想换模型就在该次调用传 `model` 参数**,不需要重连/换 session。
|
||||
|
||||
### 返回
|
||||
|
||||
工具成功时返回若干 `text` 块,第一块是摘要,后续每块各包含一张图片的 URL(方便 LLM 直接贴图或下载):
|
||||
|
||||
```jsonc
|
||||
{
|
||||
"content": [
|
||||
{ "type": "text", "text": "模型: CogView-3-Flash\n提示词: ...\n图片 1: https://maas-watermark-prod-new.cn-wlcb.ufileos.com/..." },
|
||||
{ "type": "text", "text": "https://maas-watermark-prod-new.cn-wlcb.ufileos.com/..." }
|
||||
]
|
||||
}
|
||||
```
|
||||
|
||||
URL 由智谱签名下发,**带有效期**(通常 24h),如需长期保存请在收到后及时转存到自己的存储。
|
||||
|
||||
失败时返回 `isError: true` 的错误描述。
|
||||
|
||||
---
|
||||
|
||||
## 鉴权细节
|
||||
|
||||
每个 HTTP 请求必须携带智谱 API Token,server 从以下 header 字段按顺序识别,**首个非空值生效**:
|
||||
|
||||
| 优先级 | Header | 说明 |
|
||||
|---|---|---|
|
||||
| 1 | `Authorization: Bearer <token>` | 推荐,标准 OAuth 风格 |
|
||||
| 2 | `X-Image-Gen-Token: <token>` | 直传,避免与客户端自身的 Authorization 冲突 |
|
||||
| 3 | `X-Mcp-Token: <token>` | 通用兜底 |
|
||||
|
||||
**Token 不会写入服务端任何存储**,仅在 initialize 阶段绑定到内存中的 session 对象,session 结束立即释放。
|
||||
|
||||
### 错误响应
|
||||
|
||||
| 场景 | HTTP 状态 | 表现 |
|
||||
|---|---|---|
|
||||
| initialize 时缺 token | `401` | JSON-RPC `error.message` 提示补充 header |
|
||||
| 智谱 API 4xx/5xx | `200`(MCP 工具返回) | `isError: true`,正文含智谱返回的状态码和 body |
|
||||
| 网络失败 | `200`(MCP 工具返回) | `isError: true`,正文含 `fetch` 错误信息 |
|
||||
|
||||
---
|
||||
|
||||
## 用 curl 手工调试
|
||||
|
||||
MCP Streamable HTTP 不是普通 REST,**必须按顺序**:`initialize` → `notifications/initialized` → `tools/call`。下面用一行版串起来:
|
||||
|
||||
```bash
|
||||
# 把 <your-zhipu-api-key> 替换成你智谱的 API Key
|
||||
TOKEN="<your-zhipu-api-key>"
|
||||
|
||||
# 1. initialize 拿 sessionId
|
||||
SESSION=$(curl -s -D - -X POST http://localhost:3000/mcp \
|
||||
-H "Content-Type: application/json" \
|
||||
-H "Accept: application/json, text/event-stream" \
|
||||
-H "Authorization: Bearer $TOKEN" \
|
||||
-d '{
|
||||
"jsonrpc": "2.0",
|
||||
"id": 1,
|
||||
"method": "initialize",
|
||||
"params": {
|
||||
"protocolVersion": "2024-11-05",
|
||||
"capabilities": {},
|
||||
"clientInfo": { "name": "curl", "version": "1.0" }
|
||||
}
|
||||
}' | grep -i '^mcp-session-id:' | awk '{print $2}' | tr -d '\r\n')
|
||||
|
||||
echo "Session: $SESSION"
|
||||
|
||||
# 2. notifications/initialized
|
||||
curl -s -X POST http://localhost:3000/mcp \
|
||||
-H "Content-Type: application/json" \
|
||||
-H "Accept: application/json, text/event-stream" \
|
||||
-H "Authorization: Bearer $TOKEN" \
|
||||
-H "Mcp-Session-Id: $SESSION" \
|
||||
-d '{"jsonrpc": "2.0", "method": "notifications/initialized"}'
|
||||
|
||||
# 3. 列出工具(可选验证)
|
||||
curl -s -X POST http://localhost:3000/mcp \
|
||||
-H "Content-Type: application/json" \
|
||||
-H "Accept: application/json, text/event-stream" \
|
||||
-H "Authorization: Bearer $TOKEN" \
|
||||
-H "Mcp-Session-Id: $SESSION" \
|
||||
-d '{"jsonrpc": "2.0", "id": 2, "method": "tools/list"}'
|
||||
|
||||
# 4. 调 generate_image
|
||||
curl -s -X POST http://localhost:3000/mcp \
|
||||
-H "Content-Type: application/json" \
|
||||
-H "Accept: application/json, text/event-stream" \
|
||||
-H "Authorization: Bearer $TOKEN" \
|
||||
-H "Mcp-Session-Id: $SESSION" \
|
||||
-d '{
|
||||
"jsonrpc": "2.0",
|
||||
"id": 3,
|
||||
"method": "tools/call",
|
||||
"params": {
|
||||
"name": "generate_image",
|
||||
"arguments": {
|
||||
"prompt": "一只可爱的小猫咪,坐在阳光明媚的窗台上,背景是蓝天白云",
|
||||
"size": "1280x1280"
|
||||
}
|
||||
}
|
||||
}'
|
||||
```
|
||||
|
||||
响应是 SSE 格式(`event: message` + `data: {...}`),用 `grep '^data:'` 抽 JSON 即可。
|
||||
|
||||
---
|
||||
|
||||
## 架构与文件
|
||||
|
||||
```
|
||||
image-gen/
|
||||
├── Dockerfile # node:22-slim,安装依赖后启动 MCP HTTP server
|
||||
├── .dockerignore # 排除 node_modules
|
||||
├── package.json # @modelcontextprotocol/sdk + express + zod
|
||||
├── README.md # 本文件
|
||||
└── src/
|
||||
└── mcp-server.js # 单文件实现:HTTP server + MCP transport + tool
|
||||
```
|
||||
|
||||
### 关键设计
|
||||
|
||||
- **每 session 一个 transport + server**:通过 `Mcp-Session-Id` 复用,避免每请求重建
|
||||
- **Token 闭包注入**:在 Express 入口处从 header 提取 token,闭包传给 `buildServer({ token })`,tool 回调里直接使用;不依赖 SDK 内部字段
|
||||
- **模型参数单链**:`arguments.model` → `DEFAULT_MODEL`,无 header 干预
|
||||
- **URL 提取**:`response.data[].url` 为准,若返回 `b64_json` 也会一并返回
|
||||
|
||||
### 默认模型
|
||||
|
||||
```js
|
||||
// src/mcp-server.js
|
||||
const DEFAULT_MODEL = "CogView-3-Flash";
|
||||
```
|
||||
|
||||
修改后 `docker compose up -d --build mcp-image-gen` 即可生效。
|
||||
|
||||
---
|
||||
|
||||
## 常见问题
|
||||
|
||||
**Q: 我能在工具参数里不传 `model` 吗?**
|
||||
A: 可以,不传就走默认 `CogView-3-Flash`。
|
||||
|
||||
**Q: 换模型需要重连 MCP session 吗?**
|
||||
A: 不需要。`model` 是工具参数,每次 `tools/call` 都生效,不需要重建 session。
|
||||
|
||||
**Q: Token 会不会泄露到日志?**
|
||||
A: 当前实现只在 session 内存里持有,不会写入 stdout/stderr。Container 日志只打印 session id。
|
||||
|
||||
**Q: 收到 `Missing valid Mcp-Session-Id` 错误?**
|
||||
A: 漏发了 `Mcp-Session-Id` header,或调 `tools/call` 前没先 `initialize`。
|
||||
|
||||
**Q: URL 多久过期?**
|
||||
A: 智谱签发 URL 默认 24h 左右,详见响应里的 `Expires` 参数。需要长期使用请自行转存到 OSS/S3。
|
||||
|
||||
**Q: 能否限制 `size` 白名单?**
|
||||
A: 当前透传给智谱 API,不做强制校验。如需加白名单,修改 `src/mcp-server.js` 中 `generate_image` 的 `size` schema 即可。
|
||||
@@ -25,13 +25,13 @@ function parseArgs(argv) {
|
||||
return args;
|
||||
}
|
||||
|
||||
function pickToken(req) {
|
||||
// 优先从 http header 中获取 token(兼容多种命名)
|
||||
const headers = (req && req.headers) || {};
|
||||
// 从请求头里提取 token(兼容多种命名 + Bearer 前缀)
|
||||
function extractTokenFromHeaders(headers) {
|
||||
if (!headers) return null;
|
||||
const candidates = [
|
||||
headers["authorization"],
|
||||
headers["x-image-gen-token"],
|
||||
headers["x-mcp-token"],
|
||||
headers["authorization"],
|
||||
];
|
||||
for (const v of candidates) {
|
||||
if (!v) continue;
|
||||
@@ -47,18 +47,6 @@ function pickToken(req) {
|
||||
return null;
|
||||
}
|
||||
|
||||
function pickModel(req) {
|
||||
// 从 http header 读取默认 model(可选),未设置返回 null
|
||||
const headers = (req && req.headers) || {};
|
||||
const v =
|
||||
headers["x-image-gen-model"] ||
|
||||
headers["x-mcp-model"] ||
|
||||
headers["x-model"];
|
||||
if (!v) return null;
|
||||
const s = String(v).trim();
|
||||
return s || null;
|
||||
}
|
||||
|
||||
async function callZhipuImagesGenerations({ token, model, prompt, size }) {
|
||||
const body = { model, prompt };
|
||||
if (size) body.size = size;
|
||||
@@ -99,17 +87,11 @@ function extractUrls(data) {
|
||||
return [];
|
||||
}
|
||||
|
||||
function buildServer() {
|
||||
// 创建 McpServer,把 token 通过闭包注入到工具回调里
|
||||
function buildServer({ token }) {
|
||||
const server = new McpServer(
|
||||
{
|
||||
name: "image-gen",
|
||||
version: "1.0.0",
|
||||
},
|
||||
{
|
||||
capabilities: {
|
||||
tools: {},
|
||||
},
|
||||
}
|
||||
{ name: "image-gen", version: "1.0.0" },
|
||||
{ capabilities: { tools: {} } }
|
||||
);
|
||||
|
||||
server.tool(
|
||||
@@ -131,25 +113,9 @@ function buildServer() {
|
||||
"图片尺寸,例如 1024x1024、1280x1280、1024x1792、1792x1024、768x1344、1344x768,遵循智谱 API 规格"
|
||||
),
|
||||
},
|
||||
async (args, extra) => {
|
||||
const req = extra?.requestInfo?.req;
|
||||
const token = req ? pickToken(req) : null;
|
||||
if (!token) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text:
|
||||
"缺少鉴权 token,请通过 HTTP 请求头(例如 Authorization: Bearer <token> 或 X-Image-Gen-Token: <token>)传递。",
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
|
||||
// 优先级:工具参数 model > HTTP header 中的 model > DEFAULT_MODEL
|
||||
const headerModel = req ? pickModel(req) : null;
|
||||
const model = args.model || headerModel || DEFAULT_MODEL;
|
||||
async (args) => {
|
||||
// model 优先级:工具参数 > DEFAULT_MODEL
|
||||
const model = args.model || DEFAULT_MODEL;
|
||||
const size = args.size;
|
||||
|
||||
try {
|
||||
@@ -206,17 +172,21 @@ async function startHttp({ port, host }) {
|
||||
const app = express();
|
||||
app.use(express.json({ limit: "2mb" }));
|
||||
|
||||
// session -> { server, transport } 映射
|
||||
// session -> transport + server + token/model
|
||||
const sessions = new Map();
|
||||
|
||||
app.get("/health", (_req, res) => {
|
||||
res.json({ status: "ok", service: "image-gen", sessions: sessions.size });
|
||||
});
|
||||
|
||||
async function handleSessionRequest(req, res) {
|
||||
// 主入口:先从 header 取 token(每个 HTTP 请求都取一次),
|
||||
// 然后用这个 token 创建一个新的 server+transport;
|
||||
// 已有 session 直接复用 transport,token 沿用 session 首次创建时的值
|
||||
// (避免跨请求 token 不一致;MCP 客户端本来就该每次都带同一个 token)
|
||||
const handleSessionRequest = async (req, res) => {
|
||||
const sessionId = req.headers["mcp-session-id"];
|
||||
|
||||
// 已有 session:复用 transport
|
||||
// 已有 session:复用 transport,token/model 直接从 session 缓存里拿
|
||||
if (sessionId && sessions.has(sessionId)) {
|
||||
const { transport } = sessions.get(sessionId);
|
||||
await transport.handleRequest(req, res, req.body);
|
||||
@@ -236,26 +206,42 @@ async function startHttp({ port, host }) {
|
||||
return;
|
||||
}
|
||||
|
||||
// 从本次请求 header 取 token
|
||||
const token = extractTokenFromHeaders(req.headers);
|
||||
if (!token) {
|
||||
res.status(401).json({
|
||||
jsonrpc: "2.0",
|
||||
error: {
|
||||
code: -32001,
|
||||
message:
|
||||
"缺少鉴权 token,请通过 HTTP 请求头(例如 Authorization: Bearer <token> 或 X-Image-Gen-Token: <token>)传递。",
|
||||
},
|
||||
id: req.body?.id ?? null,
|
||||
});
|
||||
return;
|
||||
}
|
||||
|
||||
const newId = randomUUID();
|
||||
const server = buildServer();
|
||||
const server = buildServer({ token });
|
||||
const transport = new StreamableHTTPServerTransport({
|
||||
sessionIdGenerator: () => newId,
|
||||
onsessioninitialized: (id) => {
|
||||
sessions.set(id, { server, transport });
|
||||
sessions.set(id, { server, transport, token });
|
||||
console.log(`[image-gen] session initialized: ${id}`);
|
||||
},
|
||||
});
|
||||
|
||||
transport.onclose = () => {
|
||||
if (transport.sessionId) {
|
||||
sessions.delete(transport.sessionId);
|
||||
console.log(`[image-gen] session closed: ${transport.sessionId}`);
|
||||
}
|
||||
};
|
||||
|
||||
await server.connect(transport);
|
||||
await transport.handleRequest(req, res, req.body);
|
||||
}
|
||||
};
|
||||
|
||||
// POST /mcp: 主入口
|
||||
app.post("/mcp", handleSessionRequest);
|
||||
|
||||
// GET /mcp & DELETE /mcp: 用于 SSE 流 / 会话关闭
|
||||
@@ -279,14 +265,14 @@ async function startHttp({ port, host }) {
|
||||
}
|
||||
|
||||
async function startStdio() {
|
||||
// 保留 stdio 启动方式,便于本地直接调试
|
||||
const server = buildServer();
|
||||
// stdio 模式下没有 HTTP header,token 暂未支持(保留入口便于本地调试)
|
||||
const server = buildServer({ token: "" });
|
||||
const { StdioServerTransport } = await import(
|
||||
"@modelcontextprotocol/sdk/server/stdio.js"
|
||||
);
|
||||
const transport = new StdioServerTransport();
|
||||
await server.connect(transport);
|
||||
console.error("[image-gen] MCP stdio server started");
|
||||
console.error("[image-gen] MCP stdio server started (no auth)");
|
||||
}
|
||||
|
||||
const args = parseArgs(process.argv.slice(2));
|
||||
|
||||
Reference in New Issue
Block a user