改名:video_gen → video_task(双工具合并为单工具,op=create/get);image_gen → image_task(单工具,op=create)
- header 前缀同步改为 X-Video-Task-* / X-Image-Task-*(方案 B,agent 侧 config.yml 的 headers 需同步更新) - docker-compose 服务名 / 容器名 / 调试端口变量(VIDEO_TASK_PORT / IMAGE_TASK_PORT)同步改名 - README(根目录与两份服务文档)按新工具形态重写:op 参数表、curl 示例、架构图
This commit is contained in:
+9
-9
@@ -5,8 +5,8 @@
|
||||
| 目录 | 容器名 | 作用 | 容器内端口 | MCP 端点 |
|
||||
|---|---|---|---|---|
|
||||
| `web_search/` | `mcp_web_search` | 免费网页搜索(Parallel / Exa,无需 key) | `3000` | `POST /` |
|
||||
| `image_gen/` | `mcp_image_gen` | 智谱 CogView 图片生成 | `3000` | `POST /mcp` |
|
||||
| `video_gen/` | `mcp_video_gen` | 智谱 CogVideo 视频生成(异步) | `3000` | `POST /mcp` |
|
||||
| `image_task/` | `mcp_image_task` | 智谱 CogView 图片生成(单工具 op=create) | `3000` | `POST /mcp` |
|
||||
| `video_task/` | `mcp_video_task` | 智谱 CogVideo 视频生成(异步,单工具 op=create/get) | `3000` | `POST /mcp` |
|
||||
|
||||
**这些服务一个宿主机端口都不占**——它们只挂在 `agent-mcp-net` 上,由 `agent-core` / `agent-executor`
|
||||
按容器名访问。所以整台服务器对外只需要开 agent-core 那一个端口(`AGENT_PORT`,默认 8090)。
|
||||
@@ -27,7 +27,7 @@ cd /root/workspace/tools/mcp && docker compose up -d --build
|
||||
> mcp 单独在一个没有该网络的机器上先起会报错(`network agent-mcp-net declared as external, but could not be found`),这是预期行为。
|
||||
> 拆的时候反着来(先 down mcp 再 down agent)。
|
||||
|
||||
`image_gen` / `video_gen` 需要智谱 token,但 token **不由服务端保存**——每个请求从 header 读取
|
||||
`image_task` / `video_task` 需要智谱 token,但 token **不由服务端保存**——每个请求从 header 读取
|
||||
(agent 侧配在 `config.yml` 的 `mcp_servers[].headers`)。所以这里不用配任何密钥。
|
||||
|
||||
### 环境变量
|
||||
@@ -37,10 +37,10 @@ cd /root/workspace/tools/mcp && docker compose up -d --build
|
||||
| 变量 | 默认 | 说明 |
|
||||
|---|---|---|
|
||||
| `WEB_SEARCH_PORT` | `8081` | 只在临时打开 web_search 调试端口时用 |
|
||||
| `IMAGE_GEN_PORT` | `8082` | 只在临时打开 image_gen 调试端口时用 |
|
||||
| `VIDEO_GEN_PORT` | `8083` | 只在临时打开 video_gen 调试端口(MCP 口,仅本机短时调试)时用 |
|
||||
| `IMAGE_TASK_PORT` | `8082` | 只在临时打开 image_task 调试端口时用 |
|
||||
| `VIDEO_TASK_PORT` | `8083` | 只在临时打开 video_task 调试端口(MCP 口,仅本机短时调试)时用 |
|
||||
|
||||
## 图片是怎么到智谱手上的(只有 video_gen 关心)
|
||||
## 图片是怎么到智谱手上的(只有 video_task 关心)
|
||||
|
||||
智谱服务器生成视频前要**自己来拉参考图**,它不带 cookie、不带 `Authorization`。这套链路里**没有任何人上传文件**:
|
||||
|
||||
@@ -48,7 +48,7 @@ cd /root/workspace/tools/mcp && docker compose up -d --build
|
||||
AI ──调用 agent 内置的「公开临时 URL」工具(直接传文件路径)──► 拿到一条短命公网 URL
|
||||
│
|
||||
▼
|
||||
AI 把拿到的 URL 当 image_url 交给 video_gen:3000/mcp
|
||||
AI 把拿到的 URL 当 image_url 交给 video_task:3000/mcp
|
||||
│
|
||||
▼
|
||||
智谱 ──匿名 GET──► agent 侧提供的公开临时 URL
|
||||
@@ -64,8 +64,8 @@ AI 把拿到的 URL 当 image_url 交给 video_gen:3000/mcp
|
||||
| 服务 | 公网暴露 | 说明 |
|
||||
|---|---|---|
|
||||
| `mcp_web_search` | 无 | 服务端无鉴权,**永远不要映射出去**(会被白嫖搜索额度);agent 走内网访问 |
|
||||
| `mcp_image_gen` | 无 | token 由调用方自带,暴露后可被拿去消耗你的智谱额度 |
|
||||
| `mcp_video_gen` | 无 | 它需要的公开临时 URL 由 agent 侧集成的工具提供,自己不占端口 |
|
||||
| `mcp_image_task` | 无 | token 由调用方自带,暴露后可被拿去消耗你的智谱额度 |
|
||||
| `mcp_video_task` | 无 | 它需要的公开临时 URL 由 agent 侧集成的工具提供,自己不占端口 |
|
||||
|
||||
## 单独调试
|
||||
|
||||
|
||||
@@ -12,23 +12,23 @@ services:
|
||||
networks:
|
||||
- agent-mcp-net
|
||||
|
||||
mcp_image_gen:
|
||||
build: ./image_gen
|
||||
container_name: mcp_image_gen
|
||||
mcp_image_task:
|
||||
build: ./image_task
|
||||
container_name: mcp_image_task
|
||||
restart: unless-stopped
|
||||
# 调试用(默认关):
|
||||
# ports:
|
||||
# - "${IMAGE_GEN_PORT:-8082}:3000"
|
||||
# - "${IMAGE_TASK_PORT:-8082}:3000"
|
||||
networks:
|
||||
- agent-mcp-net
|
||||
|
||||
mcp_video_gen:
|
||||
build: ./video_gen
|
||||
container_name: mcp_video_gen
|
||||
mcp_video_task:
|
||||
build: ./video_task
|
||||
container_name: mcp_video_task
|
||||
restart: unless-stopped
|
||||
# 调试用(默认关;这里映的是 MCP 端口,会暴露 /mcp,仅本机短时调试用):
|
||||
# ports:
|
||||
# - "${VIDEO_GEN_PORT:-8083}:3000"
|
||||
# - "${VIDEO_TASK_PORT:-8083}:3000"
|
||||
networks:
|
||||
- agent-mcp-net
|
||||
|
||||
|
||||
@@ -3,7 +3,7 @@ FROM node:22-slim
|
||||
WORKDIR /app
|
||||
|
||||
# 国内构建慢:npm 默认源在国外(web_search 零依赖不需要,这个要)。
|
||||
# 想换回官方:docker compose build --build-arg NPM_REGISTRY=https://registry.npmjs.org/ mcp_image_gen
|
||||
# 想换回官方:docker compose build --build-arg NPM_REGISTRY=https://registry.npmjs.org/ mcp_image_task
|
||||
ARG NPM_REGISTRY=https://mirrors.cloud.tencent.com/npm/
|
||||
|
||||
# 复制 package.json 并安装依赖
|
||||
@@ -1,6 +1,6 @@
|
||||
# MCP Image Gen
|
||||
# MCP Image Task
|
||||
|
||||
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。
|
||||
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。只有一个工具 `image_task`,用 `op` 区分动作(图片生成为同步接口,当前仅支持 `op=create`)。
|
||||
|
||||
- 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)
|
||||
@@ -18,17 +18,17 @@
|
||||
|
||||
```bash
|
||||
cd /root/workspace/tools/mcp
|
||||
docker compose up -d --build mcp_image_gen
|
||||
docker compose up -d --build mcp_image_task
|
||||
```
|
||||
|
||||
服务容器名 `mcp_image_gen`,容器内端口 3000,**默认不占宿主端口**:agent 走 `agent-mcp-net`
|
||||
按容器名访问(`http://mcp_image_gen:3000/mcp`)。需要从宿主机 curl 调试时,去仓库根
|
||||
`docker-compose.yml` 把 `mcp_image_gen` 注释掉的 `ports` 两行打开(默认写的是 8082),调试完记得注释回去。
|
||||
服务容器名 `mcp_image_task`,容器内端口 3000,**默认不占宿主端口**:agent 走 `agent-mcp-net`
|
||||
按容器名访问(`http://mcp_image_task:3000/mcp`)。需要从宿主机 curl 调试时,去仓库根
|
||||
`docker-compose.yml` 把 `mcp_image_task` 注释掉的 `ports` 两行打开(默认写的是 8082),调试完记得注释回去。
|
||||
|
||||
```bash
|
||||
# 不占端口的健康检查
|
||||
docker exec mcp_image_gen node -e "fetch('http://localhost:3000/health').then(r=>r.text()).then(console.log)"
|
||||
# {"status":"ok","service":"image_gen","sessions":0}
|
||||
docker exec mcp_image_task node -e "fetch('http://localhost:3000/health').then(r=>r.text()).then(console.log)"
|
||||
# {"status":"ok","service":"image_task","sessions":0}
|
||||
```
|
||||
|
||||
### MCP 客户端配置
|
||||
@@ -40,7 +40,7 @@ docker exec mcp_image_gen node -e "fetch('http://localhost:3000/health').then(r=
|
||||
// Cursor: ~/.cursor/mcp.json
|
||||
{
|
||||
"mcpServers": {
|
||||
"image_gen": {
|
||||
"image_task": {
|
||||
"url": "http://localhost:8082/mcp",
|
||||
"headers": {
|
||||
"Authorization": "Bearer <your-zhipu-api-key>"
|
||||
@@ -60,16 +60,21 @@ docker exec mcp_image_gen node -e "fetch('http://localhost:3000/health').then(r=
|
||||
|
||||
---
|
||||
|
||||
## 工具:image_gen
|
||||
## 工具:image_task
|
||||
|
||||
调用智谱图片生成接口,返回图片 URL。
|
||||
调用智谱图片生成接口,返回图片 URL。生成为**同步**接口,`op=create` 直接出结果,无需轮询。
|
||||
|
||||
### 参数
|
||||
| op | 作用 | 同步/异步 |
|
||||
|---|---|---|
|
||||
| `create` | 按 prompt 生成图片,返回图片 URL | 同步(直接返回结果) |
|
||||
|
||||
### 参数(AI 可见)
|
||||
|
||||
| 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 |
|
||||
| `size` | string | ❌ | 图片尺寸,例如 `1024x1024`、`1280x1280`、`1024x1792`、`1792x1024`、`768x1344`、`1344x768`,遵循智谱 API 规格 |
|
||||
| `op` | `"create"` | ✅ | 要执行的操作(当前仅 `create`) |
|
||||
| `prompt` | string | ✅ | 图片生成提示词,描述想要的画面内容(仅 op=create 使用) |
|
||||
| `size` | string | ❌ | 图片尺寸,例如 `1024x1024`、`1280x1280`、`1024x1792`、`1792x1024`、`768x1344`、`1344x768`,遵循智谱 API 规格(仅 op=create 使用) |
|
||||
|
||||
### Model 控制(不在工具参数中暴露)
|
||||
|
||||
@@ -77,7 +82,7 @@ docker exec mcp_image_gen node -e "fetch('http://localhost:3000/health').then(r=
|
||||
|
||||
| 优先级 | Header |
|
||||
|---|---|
|
||||
| 1 | `X-Image-Gen-Model` |
|
||||
| 1 | `X-Image-Task-Model` |
|
||||
| 2 | `X-Mcp-Model` |
|
||||
| 3 | `X-Model` |
|
||||
|
||||
@@ -87,14 +92,12 @@ docker exec mcp_image_gen node -e "fetch('http://localhost:3000/health').then(r=
|
||||
|
||||
### 返回
|
||||
|
||||
### 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": "Model: CogView-3-Flash\nPrompt: ...\nImage 1: https://maas-watermark-prod-new.cn-wlcb.ufileos.com/..." },
|
||||
{ "type": "text", "text": "https://maas-watermark-prod-new.cn-wlcb.ufileos.com/..." }
|
||||
]
|
||||
}
|
||||
@@ -112,7 +115,7 @@ URL 由智谱签名下发,**带有效期**(通常 24h),如需长期保
|
||||
|
||||
| 优先级 | Header | 说明 |
|
||||
|---|---|---|
|
||||
| 1 | `X-Image-Gen-Token: <token>` | 直传,避免与客户端自身的 Authorization 冲突 |
|
||||
| 1 | `X-Image-Task-Token: <token>` | 直传,避免与客户端自身的 Authorization 冲突 |
|
||||
| 2 | `X-Mcp-Token: <token>` | 通用兜底 |
|
||||
| 3 | `Authorization: Bearer <token>` | 标准 OAuth 风格(排最后是为避免误用) |
|
||||
|
||||
@@ -170,7 +173,7 @@ curl -s -X POST http://localhost:8082/mcp \
|
||||
-H "Mcp-Session-Id: $SESSION" \
|
||||
-d '{"jsonrpc": "2.0", "id": 2, "method": "tools/list"}'
|
||||
|
||||
# 4. 调 image_gen
|
||||
# 4. 调 image_task(op=create)
|
||||
curl -s -X POST http://localhost:8082/mcp \
|
||||
-H "Content-Type: application/json" \
|
||||
-H "Accept: application/json, text/event-stream" \
|
||||
@@ -181,8 +184,9 @@ curl -s -X POST http://localhost:8082/mcp \
|
||||
"id": 3,
|
||||
"method": "tools/call",
|
||||
"params": {
|
||||
"name": "image_gen",
|
||||
"name": "image_task",
|
||||
"arguments": {
|
||||
"op": "create",
|
||||
"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"
|
||||
}
|
||||
@@ -197,20 +201,21 @@ curl -s -X POST http://localhost:8082/mcp \
|
||||
## 架构与文件
|
||||
|
||||
```
|
||||
image_gen/
|
||||
image_task/
|
||||
├── 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
|
||||
└── mcp-server.js # 单文件实现:HTTP server + MCP transport + 单工具(op 分发)
|
||||
```
|
||||
|
||||
### 关键设计
|
||||
|
||||
- **单工具 + `op` 分发**:`image_task(op=create)`,与 `video_task` 的服务形态保持一致;生成为同步接口,所以 `op` 只有 `create` 一个取值
|
||||
- **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
|
||||
- **Token 闭包注入**:在 Express 入口处从 header 提取 token,闭包传给 tool 回调;不依赖 SDK 内部字段
|
||||
- **模型参数不入 schema**:model 只从 HTTP header 读取(`X-Image-Task-Model` / `X-Mcp-Model` / `X-Model`),AI 不可见
|
||||
- **URL 提取**:`response.data[].url` 为准,若返回 `b64_json` 也会一并返回
|
||||
|
||||
### 默认模型
|
||||
@@ -220,18 +225,21 @@ image_gen/
|
||||
const DEFAULT_MODEL = "CogView-3-Flash";
|
||||
```
|
||||
|
||||
修改后 `docker compose up -d --build mcp_image_gen` 即可生效。
|
||||
修改后 `docker compose up -d --build mcp_image_task` 即可生效。
|
||||
|
||||
---
|
||||
|
||||
## 常见问题
|
||||
|
||||
**Q: AI 不知道有哪些 model 可选,怎么办?**
|
||||
A: 这是设计意图。模型选择属于路由策略,不让 AI 决策。调用方在 HTTP header `X-Image-Gen-Model` 里注入即可。
|
||||
A: 这是设计意图。模型选择属于路由策略,不让 AI 决策。调用方在 HTTP header `X-Image-Task-Model` 里注入即可。
|
||||
|
||||
**Q: 换模型需要重连 MCP session 吗?**
|
||||
A: 不需要。Model header 在每次 `tools/call` 都从当前请求读取,不需要重建 session。
|
||||
|
||||
**Q: 为什么 op 只有 create?**
|
||||
A: 图片生成是同步接口,`op=create` 直接返回结果,不存在需要轮询的异步任务。保留 `op` 参数是为了与 `video_task` 等异步任务型服务形态一致;将来若上游出现异步接口再加 `get`。
|
||||
|
||||
**Q: 为什么 Model 优先级不是「工具参数 > header」?**
|
||||
A: 因为我们故意不在 schema 暴露 model 参数,AI 根本不会传 model,所以「工具参数 > header」这条分支实际上不存在。
|
||||
|
||||
@@ -245,4 +253,4 @@ A: 漏发了 `Mcp-Session-Id` header,或调 `tools/call` 前没先 `initialize
|
||||
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 `image_gen` 的 `size` schema 即可。
|
||||
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 `image_task` 的 `size` schema 即可。
|
||||
@@ -1,7 +1,7 @@
|
||||
{
|
||||
"name": "mcp_image_gen",
|
||||
"name": "mcp_image_task",
|
||||
"version": "1.0.0",
|
||||
"description": "MCP server for image generation via Zhipu BigModel CogView-3-Flash",
|
||||
"description": "MCP server for image generation via Zhipu BigModel CogView (single tool image_task with op=create)",
|
||||
"type": "module",
|
||||
"main": "src/mcp-server.js",
|
||||
"scripts": {
|
||||
@@ -13,7 +13,7 @@ const DEFAULT_MODEL = "CogView-3-Flash";
|
||||
// ── 优雅退出 ──────────────────────────────────────────────────────────────
|
||||
// 容器里 node 以 PID 1 运行且未注册 SIGTERM 处理器时,内核不会对其执行默认
|
||||
// 终止动作,docker stop/restart 会白等满 10s 宽限期才被 SIGKILL。
|
||||
// 显式注册处理器,关闭 MCP session 与长连接后立即退出;3s 定时器为兑底。
|
||||
// 显式注册处理器,关闭 MCP session 与长连接后立即退出;3s 定时器为兜底。
|
||||
let httpServer = null;
|
||||
let closeAllSessions = () => {};
|
||||
let shuttingDown = false;
|
||||
@@ -21,7 +21,7 @@ let shuttingDown = false;
|
||||
function shutdown(sig) {
|
||||
if (shuttingDown) return;
|
||||
shuttingDown = true;
|
||||
console.log(`[image_gen] 收到 ${sig},停止服务`);
|
||||
console.log(`[image] received ${sig}, shutting down`);
|
||||
closeAllSessions();
|
||||
if (!httpServer) process.exit(0); // stdio 模式无监听,直接退
|
||||
httpServer.close(() => process.exit(0));
|
||||
@@ -48,7 +48,7 @@ function parseArgs(argv) {
|
||||
function pickToken(headers) {
|
||||
if (!headers) return null;
|
||||
const candidates = [
|
||||
headers["x-image-gen-token"],
|
||||
headers["x-image-task-token"],
|
||||
headers["x-mcp-token"],
|
||||
headers["authorization"],
|
||||
];
|
||||
@@ -69,7 +69,7 @@ function pickToken(headers) {
|
||||
function pickModel(headers) {
|
||||
if (!headers) return null;
|
||||
const v =
|
||||
headers["x-image-gen-model"] ||
|
||||
headers["x-image-task-model"] ||
|
||||
headers["x-mcp-model"] ||
|
||||
headers["x-model"];
|
||||
if (!v) return null;
|
||||
@@ -117,10 +117,64 @@ function extractUrls(data) {
|
||||
return [];
|
||||
}
|
||||
|
||||
// ===== op=create:提交图片生成(同步,直接返回图片 URL)=====
|
||||
async function handleCreate({ args, headers, token }) {
|
||||
// model 完全由 HTTP header 控制(X-Image-Task-Model / X-Mcp-Model / X-Model),
|
||||
// 不存进 session,每次从本次请求的 header 读取,未设置则用默认 CogView-3-Flash。
|
||||
// AI 看不到 model 参数(schema 中已删除),避免其误传。
|
||||
const model = pickModel(headers) || DEFAULT_MODEL;
|
||||
const size = args.size;
|
||||
|
||||
try {
|
||||
const data = await callZhipuImagesGenerations({
|
||||
token,
|
||||
model,
|
||||
prompt: args.prompt,
|
||||
size,
|
||||
});
|
||||
const urls = extractUrls(data);
|
||||
if (urls.length === 0) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Returned no image URL. Raw response: ${JSON.stringify(data)}`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
const summary = urls
|
||||
.map((u, i) => `Image ${i + 1}: ${u}`)
|
||||
.join("\n");
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Model: ${model}\nPrompt: ${args.prompt}\n${summary}`,
|
||||
},
|
||||
...urls.map((u) => ({ type: "text", text: u })),
|
||||
],
|
||||
};
|
||||
} catch (e) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Failed to generate image: ${e.message}${
|
||||
e.detail ? `\nDetails: ${JSON.stringify(e.detail)}` : ""
|
||||
}`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
}
|
||||
|
||||
function buildServer() {
|
||||
const server = new McpServer(
|
||||
{
|
||||
name: "image_gen",
|
||||
name: "image_task",
|
||||
version: "1.0.0",
|
||||
},
|
||||
{
|
||||
@@ -131,18 +185,27 @@ function buildServer() {
|
||||
);
|
||||
|
||||
server.tool(
|
||||
"image_gen",
|
||||
"Generate an image from a prompt and return the image URL. Optionally specify size.",
|
||||
"image_task",
|
||||
`Manages image generation. Generation is synchronous: op=create returns the image URLs directly.
|
||||
|
||||
op=create Generate an image from a prompt. Requires prompt; optional size.`,
|
||||
{
|
||||
op: z
|
||||
.enum(["create"])
|
||||
.describe(
|
||||
'Operation to perform. Only "create" is supported (generation is synchronous, so there is no task to poll).'
|
||||
),
|
||||
prompt: z
|
||||
.string()
|
||||
.min(1)
|
||||
.describe("Image generation prompt; describe the desired scene in detail."),
|
||||
.describe(
|
||||
"Image generation prompt; describe the desired scene in detail. Used with op=create."
|
||||
),
|
||||
size: z
|
||||
.string()
|
||||
.optional()
|
||||
.describe(
|
||||
"Image size, Preferred sizes: 1024x1024, 1280x1280, 1536x1024, 1024x1536, Each side must be a multiple of 32."
|
||||
"Image size, Preferred sizes: 1024x1024, 1280x1280, 1536x1024, 1024x1536, Each side must be a multiple of 32. Used with op=create."
|
||||
),
|
||||
},
|
||||
async (args, extra) => {
|
||||
@@ -163,55 +226,19 @@ function buildServer() {
|
||||
};
|
||||
}
|
||||
|
||||
// model 完全由 HTTP header 控制(X-Image-Gen-Model / X-Mcp-Model / X-Model),
|
||||
// 不存进 session,每次从本次请求的 header 读取,未设置则用默认 CogView-3-Flash。
|
||||
// AI 看不到 model 参数(schema 中已删除),避免其误传。
|
||||
const model = pickModel(headers) || DEFAULT_MODEL;
|
||||
const size = args.size;
|
||||
|
||||
try {
|
||||
const data = await callZhipuImagesGenerations({
|
||||
token,
|
||||
model,
|
||||
prompt: args.prompt,
|
||||
size,
|
||||
});
|
||||
const urls = extractUrls(data);
|
||||
if (urls.length === 0) {
|
||||
switch (args.op) {
|
||||
case "create":
|
||||
return await handleCreate({ args, headers, token });
|
||||
default:
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Returned no image URL. Raw response: ${JSON.stringify(data)}`,
|
||||
text: `Unknown op "${args.op}" (available: create).`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
const summary = urls
|
||||
.map((u, i) => `Image ${i + 1}: ${u}`)
|
||||
.join("\n");
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Model: ${model}\nPrompt: ${args.prompt}\n${summary}`,
|
||||
},
|
||||
...urls.map((u) => ({ type: "text", text: u })),
|
||||
],
|
||||
};
|
||||
} catch (e) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Failed to generate image: ${e.message}${
|
||||
e.detail ? `\nDetails: ${JSON.stringify(e.detail)}` : ""
|
||||
}`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
}
|
||||
);
|
||||
@@ -241,7 +268,7 @@ async function startHttp({ port, host }) {
|
||||
};
|
||||
|
||||
app.get("/health", (_req, res) => {
|
||||
res.json({ status: "ok", service: "image_gen", sessions: sessions.size });
|
||||
res.json({ status: "ok", service: "image_task", sessions: sessions.size });
|
||||
});
|
||||
|
||||
async function handleSessionRequest(req, res) {
|
||||
@@ -304,7 +331,7 @@ async function startHttp({ port, host }) {
|
||||
|
||||
httpServer = app.listen(port, host, () => {
|
||||
console.log(
|
||||
`[image_gen] MCP HTTP server listening on http://${host}:${port}/mcp`
|
||||
`[image] MCP HTTP server listening on http://${host}:${port}/mcp`
|
||||
);
|
||||
});
|
||||
}
|
||||
@@ -317,18 +344,18 @@ async function startStdio() {
|
||||
);
|
||||
const transport = new StdioServerTransport();
|
||||
await server.connect(transport);
|
||||
console.error("[image_gen] MCP stdio server started");
|
||||
console.error("[image] MCP stdio server started");
|
||||
}
|
||||
|
||||
const args = parseArgs(process.argv.slice(2));
|
||||
if (args.http) {
|
||||
startHttp({ port: args.port, host: args.host }).catch((e) => {
|
||||
console.error("[image_gen] failed to start http server:", e);
|
||||
console.error("[image] failed to start http server:", e);
|
||||
process.exit(1);
|
||||
});
|
||||
} else {
|
||||
startStdio().catch((e) => {
|
||||
console.error("[image_gen] failed to start stdio server:", e);
|
||||
console.error("[image] failed to start stdio server:", e);
|
||||
process.exit(1);
|
||||
});
|
||||
}
|
||||
@@ -3,7 +3,7 @@ FROM node:22-slim
|
||||
WORKDIR /app
|
||||
|
||||
# 国内构建慢:npm 默认源在国外。想换回官方:
|
||||
# docker compose build --build-arg NPM_REGISTRY=https://registry.npmjs.org/ mcp_video_gen
|
||||
# docker compose build --build-arg NPM_REGISTRY=https://registry.npmjs.org/ mcp_video_task
|
||||
ARG NPM_REGISTRY=https://mirrors.cloud.tencent.com/npm/
|
||||
|
||||
# 复制 package.json 并安装依赖
|
||||
@@ -16,4 +16,4 @@ COPY . .
|
||||
EXPOSE 3000
|
||||
|
||||
# HTTP 模式:只提供 MCP(/mcp)与 /health;容器外不映射,只在 agent-mcp-net 内被 agent 访问。
|
||||
CMD ["node", "src/mcp-server.js", "--http", "--port", "3000", "--host", "0.0.0.0"]
|
||||
CMD ["node", "src/mcp-server.js", "--http", "--port", "3000", "--host", "0.0.0.0"]
|
||||
@@ -1,6 +1,6 @@
|
||||
# MCP Video Gen
|
||||
# MCP Video Task
|
||||
|
||||
基于 [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) 的视频生成服务,对接智谱 BigModel 的视频生成**异步**接口,支持**文生视频**和**图生视频**。Server 只负责提交任务 + 单次查询结果,**不内部轮询**——AI 拿到 `task_id` 后自己决定什么时候调用 `query_video` 复查。
|
||||
基于 [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) 的视频生成服务,对接智谱 BigModel 的视频生成**异步**接口,支持**文生视频**和**图生视频**。只有一个工具 `video_task`,用 `op` 区分动作:`op=create` 提交任务、`op=get` 单次查询结果。Server **不内部轮询**——AI 拿到 `task_id` 后自己决定什么时候用 `op="get"` 复查。
|
||||
|
||||
- 协议:MCP **Streamable HTTP**(单端点 `/mcp`,支持 SSE 流)
|
||||
- 上游:[智谱 BigModel 视频生成(异步)](https://docs.bigmodel.cn/api-reference/模型-api/视频生成异步) + [查询异步结果](https://docs.bigmodel.cn/api-reference/模型-api/查询异步结果)
|
||||
@@ -19,17 +19,17 @@
|
||||
|
||||
```bash
|
||||
cd /root/workspace/tools/mcp
|
||||
docker compose up -d --build mcp_video_gen
|
||||
docker compose up -d --build mcp_video_task
|
||||
```
|
||||
|
||||
服务容器名 `mcp_video_gen`,**默认不占宿主机端口**:由 agent 走 `agent-mcp-net` 按容器名访问
|
||||
(`http://mcp_video_gen:3000/mcp`)。需要从宿主机 curl 调试时,去仓库根 `docker-compose.yml`
|
||||
把 `mcp_video_gen` 注释掉的 `ports` 两行打开(映的是 MCP 端口 3000,仅本机短时调试用)。
|
||||
服务容器名 `mcp_video_task`,**默认不占宿主机端口**:由 agent 走 `agent-mcp-net` 按容器名访问
|
||||
(`http://mcp_video_task:3000/mcp`)。需要从宿主机 curl 调试时,去仓库根 `docker-compose.yml`
|
||||
把 `mcp_video_task` 注释掉的 `ports` 两行打开(映的是 MCP 端口 3000,仅本机短时调试用)。
|
||||
|
||||
```bash
|
||||
# 容器内健康检查(无需占端口)
|
||||
docker exec mcp_video_gen node -e "fetch('http://localhost:3000/health').then(r=>r.text()).then(console.log)"
|
||||
# {"status":"ok","service":"video_gen","sessions":0}
|
||||
docker exec mcp_video_task node -e "fetch('http://localhost:3000/health').then(r=>r.text()).then(console.log)"
|
||||
# {"status":"ok","service":"video_task","sessions":0}
|
||||
```
|
||||
|
||||
> 下文 curl 示例里的 `localhost:8083` 均指“临时打开了调试端口”的情形。
|
||||
@@ -43,7 +43,7 @@ docker exec mcp_video_gen node -e "fetch('http://localhost:3000/health').then(r=
|
||||
// Cursor: ~/.cursor/mcp.json
|
||||
{
|
||||
"mcpServers": {
|
||||
"video_gen": {
|
||||
"video_task": {
|
||||
"url": "http://localhost:8083/mcp",
|
||||
"headers": {
|
||||
"Authorization": "Bearer <your-zhipu-api-key>"
|
||||
@@ -59,77 +59,51 @@ docker exec mcp_video_gen node -e "fetch('http://localhost:3000/health').then(r=
|
||||
|
||||
---
|
||||
|
||||
## 工具一览
|
||||
## 工具:video_task
|
||||
|
||||
| 工具 | 作用 | 同步/异步 |
|
||||
单个工具,`op` 必填,只有两个取值:
|
||||
|
||||
| op | 作用 | 同步/异步 |
|
||||
|---|---|---|
|
||||
| `video_gen` | 提交一个视频生成任务,立刻返回 `task_id` | 同步(仅提交) |
|
||||
| `query_video` | 用 `task_id` 拉一次当前结果 | 同步(单次查询,不轮询) |
|
||||
|
||||
---
|
||||
|
||||
|
||||
## 本地文件 → 公网 URL(本服务不搬字节)
|
||||
|
||||
`image_url` 必须是一个**公网可达**的 URL(智谱服务器会自己去拉)。图片在本地盘上时,本服务不负责上传:
|
||||
agent 侧已集成为一个 **tool**,AI 直接传文件路径就能拿到一条公开临时 URL,拿来填 `image_url` 即可,
|
||||
不需要 curl 任何地址、不需要任何配置。
|
||||
|
||||
- 文件始终躺在 agent 的任务目录里,零拷贝;链接短命、到期或 agent 重启后失效。
|
||||
- 智谱只在提交任务那一刻拉一次图,短命链接完全够用。
|
||||
|
||||
---
|
||||
|
||||
## 工具:video_gen
|
||||
|
||||
提交一个视频生成任务,支持**文生视频**和**图生视频**。`prompt` 为**必填**;图生视频另传 `image_url`(两者都传也行,此时 prompt 会附加为运动/动作描述)。
|
||||
| `create` | 提交一个视频生成任务,立刻返回 `task_id` | 同步(仅提交) |
|
||||
| `get` | 用 `task_id` 拉一次当前结果 | 同步(单次查询,不轮询) |
|
||||
|
||||
### 参数(AI 可见)
|
||||
|
||||
| 名称 | 类型 | 必填 | 说明 |
|
||||
|---|---|---|---|
|
||||
| `prompt` | string | ✅ | 视频文本描述,最大 512 字符 |
|
||||
| `op` | `"create"` \| `"get"` | ✅ | 要执行的操作 |
|
||||
| `prompt` | string | op=create 时与 `image_url` 二选一 | 视频文本描述,最大 512 字符;与 `image_url` 同传时作为运动/动作描述 |
|
||||
| `image_url` | string (URL) | ❌ | 参考图 URL(**单张**,不支持数组 / 多张 / 首尾帧)。图片 ≤ 5MB,格式 png/jpeg/jpg。本地文件请先用 agent 内置的公开临时 URL 工具换成公网链接(上一节) |
|
||||
| `with_audio` | boolean | ❌ | 是否生成 AI 音效,默认 `false` |
|
||||
| `with_audio` | boolean | ❌ | 是否生成 AI 音效,默认 `false`(仅 op=create 使用) |
|
||||
| `task_id` | string | op=get 时必填 | `op=create` 返回的 `id` |
|
||||
|
||||
### 服务端 Header 覆盖项(AI 看不到)
|
||||
|
||||
| Header | 类型 | 默认 | 说明 |
|
||||
|---|---|---|---|
|
||||
| `X-Video-Gen-Model` / `X-Mcp-Model` / `X-Model` | string | `cogvideox-flash` | 覆盖模型 |
|
||||
| `X-Video-Gen-Quality` | string | `speed` | `speed`(速度优先)或 `quality`(质量优先) |
|
||||
| `X-Video-Gen-Fps` | int | `30` | `30` 或 `60` |
|
||||
| `X-Video-Gen-Watermark` | bool | `true` | 是否添加 AI 水印;关闭需先在智谱后台签署免责声明 |
|
||||
| `X-Video-Task-Model` / `X-Mcp-Model` / `X-Model` | string | `cogvideox-flash` | 覆盖模型 |
|
||||
| `X-Video-Task-Quality` | string | `speed` | `speed`(速度优先)或 `quality`(质量优先) |
|
||||
| `X-Video-Task-Fps` | int | `30` | `30` 或 `60` |
|
||||
| `X-Video-Task-Watermark` | bool | `true` | 是否添加 AI 水印;关闭需先在智谱后台签署免责声明 |
|
||||
|
||||
模型名串识别顺序(首个非空值生效):`X-Video-Gen-Model` → `X-Mcp-Model` → `X-Model` → 默认。
|
||||
模型名串识别顺序(首个非空值生效):`X-Video-Task-Model` → `X-Mcp-Model` → `X-Model` → 默认。
|
||||
|
||||
### 返回
|
||||
### 返回(op=create)
|
||||
|
||||
工具立刻返回智谱提交结果:
|
||||
立刻返回智谱提交结果:
|
||||
|
||||
```jsonc
|
||||
{
|
||||
"content": [
|
||||
{ "type": "text", "text": "Model: cogvideox-flash\ntask_id: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx\nrequest_id: yyyyyyyy-yyyy-yyyy-yyyy-yyyyyyyyyyyy\ntask_status: PROCESSING\n\nVideo generation is asynchronous. Call query_video with task_id=\"xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx\" to fetch the result." }
|
||||
{ "type": "text", "text": "Model: cogvideox-flash\ntask_id: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx\nrequest_id: yyyyyyyy-yyyy-yyyy-yyyy-yyyyyyyyyyyy\ntask_status: PROCESSING\n\nVideo generation is asynchronous. Call this tool again with op=\"get\" and task_id=\"xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx\" to fetch the result." }
|
||||
]
|
||||
}
|
||||
```
|
||||
|
||||
`task_status` 通常为 `PROCESSING`,极少数情况下可能直接 `SUCCESS`(极快完成的短任务)。
|
||||
|
||||
---
|
||||
|
||||
## 工具:query_video
|
||||
|
||||
根据 `task_id` 拉一次智谱异步结果。**不内部轮询**——AI 自己决定调几次、调多频。
|
||||
|
||||
### 参数(AI 可见)
|
||||
|
||||
| 名称 | 类型 | 必填 | 说明 |
|
||||
|---|---|---|---|
|
||||
| `task_id` | string | ✅ | `video_gen` 返回的 `id` |
|
||||
|
||||
### 返回
|
||||
### 返回(op=get)
|
||||
|
||||
- **`SUCCESS`**:返回 `Video URL`(智谱签名 URL,**24h 左右有效**,需及时转存),外加一张 `Cover` 封面图:
|
||||
```jsonc
|
||||
@@ -145,13 +119,24 @@ agent 侧已集成为一个 **tool**,AI 直接传文件路径就能拿到一
|
||||
|
||||
---
|
||||
|
||||
## 本地文件 → 公网 URL(本服务不搬字节)
|
||||
|
||||
`image_url` 必须是一个**公网可达**的 URL(智谱服务器会自己去拉)。图片在本地盘上时,本服务不负责上传:
|
||||
agent 侧已集成为一个 **tool**,AI 直接传文件路径就能拿到一条公开临时 URL,拿来填 `image_url` 即可,
|
||||
不需要 curl 任何地址、不需要任何配置。
|
||||
|
||||
- 文件始终躺在 agent 的任务目录里,零拷贝;链接短命、到期或 agent 重启后失效。
|
||||
- 智谱只在提交任务那一刻拉一次图,短命链接完全够用。
|
||||
|
||||
---
|
||||
|
||||
## 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 | `X-Video-Gen-Token: <token>` | 直传,避免与客户端自身的 Authorization 冲突 |
|
||||
| 1 | `X-Video-Task-Token: <token>` | 直传,避免与客户端自身的 Authorization 冲突 |
|
||||
| 2 | `X-Mcp-Token: <token>` | 通用兜底 |
|
||||
| 3 | `Authorization: Bearer <token>` | 标准 OAuth 风格(排最后是为避免误用) |
|
||||
|
||||
@@ -171,9 +156,9 @@ Token 仅在每次 tool call 时从 header 读取,不写入任何存储、不
|
||||
|
||||
```
|
||||
1. 如果手头是【本地文件】:先调用 agent 内置的「公开临时 URL」工具拿到一条公网链接(上一节),再进第 2 步
|
||||
2. 调用 video_gen({ prompt, image_url?, with_audio? })
|
||||
2. 调用 video_task({ op: "create", prompt, image_url?, with_audio? })
|
||||
3. 拿到 task_id → 等待若干秒(建议 2~5 秒)
|
||||
4. 调用 query_video({ task_id })
|
||||
4. 调用 video_task({ op: "get", task_id })
|
||||
- 若 PROCESSING → 等一会再调
|
||||
- 若 SUCCESS → 拿到 video URL,贴给用户 / 下载 / 二次编辑
|
||||
- 若 FAIL → 展示错误信息,结束
|
||||
@@ -225,7 +210,7 @@ curl -s -X POST http://localhost:8083/mcp \
|
||||
-H "Mcp-Session-Id: $SESSION" \
|
||||
-d '{"jsonrpc": "2.0", "id": 2, "method": "tools/list"}'
|
||||
|
||||
# 4. 提交文生视频任务
|
||||
# 4. 提交文生视频任务(op=create)
|
||||
curl -s -X POST http://localhost:8083/mcp \
|
||||
-H "Content-Type: application/json" \
|
||||
-H "Accept: application/json, text/event-stream" \
|
||||
@@ -236,15 +221,16 @@ curl -s -X POST http://localhost:8083/mcp \
|
||||
"id": 3,
|
||||
"method": "tools/call",
|
||||
"params": {
|
||||
"name": "video_gen",
|
||||
"name": "video_task",
|
||||
"arguments": {
|
||||
"op": "create",
|
||||
"prompt": "A cat is playing with a ball.",
|
||||
"with_audio": true
|
||||
}
|
||||
}
|
||||
}'
|
||||
|
||||
# 5. 查结果(替换下面的 <task_id>)
|
||||
# 5. 查结果(op=get,替换下面的 <task_id>)
|
||||
curl -s -X POST http://localhost:8083/mcp \
|
||||
-H "Content-Type: application/json" \
|
||||
-H "Accept: application/json, text/event-stream" \
|
||||
@@ -255,8 +241,8 @@ curl -s -X POST http://localhost:8083/mcp \
|
||||
"id": 4,
|
||||
"method": "tools/call",
|
||||
"params": {
|
||||
"name": "query_video",
|
||||
"arguments": { "task_id": "<task_id>" }
|
||||
"name": "video_task",
|
||||
"arguments": { "op": "get", "task_id": "<task_id>" }
|
||||
}
|
||||
}'
|
||||
```
|
||||
@@ -266,23 +252,24 @@ curl -s -X POST http://localhost:8083/mcp \
|
||||
## 架构与文件
|
||||
|
||||
```
|
||||
video_gen/
|
||||
video_task/
|
||||
├── Dockerfile # node:22-slim,安装依赖后启动 MCP HTTP server
|
||||
├── .dockerignore # 排除 node_modules
|
||||
package.json # @modelcontextprotocol/sdk + express + multer
|
||||
package.json # @modelcontextprotocol/sdk + express
|
||||
├── README.md # 本文件
|
||||
└── src/
|
||||
└── mcp-server.js # 单文件实现:HTTP server + MCP transport + 双工具
|
||||
└── mcp-server.js # 单文件实现:HTTP server + MCP transport + 单工具(op 分发)
|
||||
```
|
||||
|
||||
### 关键设计
|
||||
|
||||
- **单工具 + `op` 分发**:`video_task(op=create|get)` 替代原来的 `video_gen` + `query_video` 双工具,AI 面对的工具面更小,参数按 op 在 handler 侧校验
|
||||
- **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 tool 回调;不依赖 SDK 内部字段
|
||||
- **AI 可见 vs AI 不可见**:
|
||||
- AI 只看到 `prompt / image_url / with_audio`
|
||||
- AI 只看到 `op / prompt / image_url / with_audio / task_id`
|
||||
- `model / quality / fps / watermark_enabled` 全部由 HTTP header 注入,避免 AI 误传或被诱导调高价档
|
||||
- **异步语义暴露给 AI**:服务端不做轮询,`video_gen` 只提交、`query_video` 只查一次。AI 自主决定节奏
|
||||
- **异步语义暴露给 AI**:服务端不做轮询,`op=create` 只提交、`op=get` 只查一次。AI 自主决定节奏
|
||||
- **服务端自动 `request_id`**:每次提交服务端生成一个 UUID 作为智谱 `request_id`,便于日志关联,但不暴露给 AI
|
||||
- **不搬字节**:本地文件 -> 公网 URL 完全由 agent 侧集成的「公开临时 URL」工具完成,本服务不参与;所以本体不加任何写入口、不落盘、不需要额外端口
|
||||
- **链接短命**:公开临时 URL 到期或 agent 重启后失效;智谱只需在提交任务那一刻拉一次图
|
||||
@@ -294,13 +281,13 @@ package.json # @modelcontextprotocol/sdk + express + multer
|
||||
const DEFAULT_MODEL = "cogvideox-flash";
|
||||
```
|
||||
|
||||
修改后 `docker compose up -d --build mcp_video_gen` 即可生效。
|
||||
修改后 `docker compose up -d --build mcp_video_task` 即可生效。
|
||||
|
||||
---
|
||||
|
||||
## 常见问题
|
||||
|
||||
**Q: AI 拿到 task_id 后立刻调 query_video 一直 PROCESSING,怎么办?**
|
||||
**Q: AI 拿到 task_id 后立刻 op=get 一直 PROCESSING,怎么办?**
|
||||
A: AI 自己 sleep 几秒再调(建议 2~5 秒间隔)。这是设计取舍:把节奏控制权交给 AI 模型,而不是服务端硬编码超时。
|
||||
|
||||
**Q: 视频 URL 多久过期?**
|
||||
@@ -319,4 +306,4 @@ A: 可以。`/async-result/{id}` 是幂等的读接口,并发无副作用。
|
||||
A: 不需要。Model header 在每次 `tools/call` 都从当前请求读取。
|
||||
|
||||
**Q: 上传的图片多久过期?**
|
||||
A: 本服务不存图,不存在“上传的图片”:要用的图在 agent 的任务目录里,AI 用 agent 内置的公开临时 URL 工具现取一条短命链接给智谱拉。链接过期或 agent 重启后该 URL 失效,但智谱只在提交任务时拉一次,不受影响。
|
||||
A: 本服务不存图,不存在“上传的图片”:要用的图在 agent 的任务目录里,AI 用 agent 内置的公开临时 URL 工具现取一条短命链接给智谱拉。链接过期或 agent 重启后该 URL 失效,但智谱只在提交任务时拉一次,不受影响。
|
||||
@@ -1,7 +1,7 @@
|
||||
{
|
||||
"name": "mcp_video_gen",
|
||||
"name": "mcp_video_task",
|
||||
"version": "1.0.0",
|
||||
"description": "MCP server for video generation via Zhipu BigModel cogvideox-flash (text-to-video / image-to-video, async)",
|
||||
"description": "MCP server for async video generation tasks via Zhipu BigModel cogvideox-flash (single tool video_task with op=create/get)",
|
||||
"type": "module",
|
||||
"main": "src/mcp-server.js",
|
||||
"scripts": {
|
||||
@@ -11,18 +11,18 @@ const ZHIPU_SUBMIT_URL =
|
||||
const ZHIPU_QUERY_URL = (id) =>
|
||||
`https://open.bigmodel.cn/api/paas/v4/async-result/${encodeURIComponent(id)}`;
|
||||
|
||||
// 默认模型:你说的 cogvideox-flash,文档枚举里也确认可用
|
||||
// 默认模型:cogvideox-flash
|
||||
const DEFAULT_MODEL = "cogvideox-flash";
|
||||
|
||||
// AI 不可见的服务端默认(可被同名 header 覆盖)
|
||||
const DEFAULT_QUALITY = "speed"; // speed | quality
|
||||
const DEFAULT_FPS = 30; // 30 | 60
|
||||
const DEFAULT_FPS = 30; // 30 | 60
|
||||
const DEFAULT_WATERMARK_ENABLED = true; // 平台政策默认 true
|
||||
|
||||
// ── 优雅退出 ──────────────────────────────────────────────────────────────
|
||||
// 容器里 node 以 PID 1 运行且未注册 SIGTERM 处理器时,内核不会对其执行默认
|
||||
// 终止动作,docker stop/restart 会白等满 10s 宽限期才被 SIGKILL。
|
||||
// 显式注册处理器,关闭 MCP session 与长连接后立即退出;3s 定时器为兑底。
|
||||
// 显式注册处理器,关闭 MCP session 与长连接后立即退出;3s 定时器为兜底。
|
||||
let httpServer = null;
|
||||
let closeAllSessions = () => {};
|
||||
let shuttingDown = false;
|
||||
@@ -30,7 +30,7 @@ let shuttingDown = false;
|
||||
function shutdown(sig) {
|
||||
if (shuttingDown) return;
|
||||
shuttingDown = true;
|
||||
console.log(`[video_gen] 收到 ${sig},停止服务`);
|
||||
console.log(`[video] received ${sig}, shutting down`);
|
||||
closeAllSessions();
|
||||
if (!httpServer) process.exit(0); // stdio 模式无监听,直接退
|
||||
httpServer.close(() => process.exit(0));
|
||||
@@ -57,7 +57,7 @@ function parseArgs(argv) {
|
||||
function pickToken(headers) {
|
||||
if (!headers) return null;
|
||||
const candidates = [
|
||||
headers["x-video-gen-token"],
|
||||
headers["x-video-task-token"],
|
||||
headers["x-mcp-token"],
|
||||
headers["authorization"],
|
||||
];
|
||||
@@ -78,7 +78,7 @@ function pickToken(headers) {
|
||||
function pickModel(headers) {
|
||||
if (!headers) return null;
|
||||
const v =
|
||||
headers["x-video-gen-model"] ||
|
||||
headers["x-video-task-model"] ||
|
||||
headers["x-mcp-model"] ||
|
||||
headers["x-model"];
|
||||
if (!v) return null;
|
||||
@@ -140,7 +140,7 @@ async function callZhipu({ url, token, method = "POST", body }) {
|
||||
return data;
|
||||
}
|
||||
|
||||
// ===== 工具:提交视频生成(同步只提交,立刻返回)=====
|
||||
// ===== op=create:提交视频生成(同步只提交,立刻返回)=====
|
||||
async function submitVideo({
|
||||
token,
|
||||
model,
|
||||
@@ -151,7 +151,7 @@ async function submitVideo({
|
||||
fps,
|
||||
watermark_enabled,
|
||||
}) {
|
||||
// 平台要求 image_url 和 prompt 至少传一个
|
||||
// 平台要求 image_url 和 prompt 至少传一个(handler 侧已校验)
|
||||
const body = {
|
||||
model,
|
||||
quality,
|
||||
@@ -164,143 +164,202 @@ async function submitVideo({
|
||||
if (prompt) body.prompt = prompt;
|
||||
if (image_url) body.image_url = image_url;
|
||||
|
||||
const data = await callZhipu({
|
||||
url: ZHIPU_SUBMIT_URL,
|
||||
token,
|
||||
body,
|
||||
});
|
||||
return data; // { model, id, request_id, task_status }
|
||||
return await callZhipu({ url: ZHIPU_SUBMIT_URL, token, body });
|
||||
// { model, id, request_id, task_status }
|
||||
}
|
||||
|
||||
// ===== 工具:查询异步结果(单次拉取,不内部轮询)=====
|
||||
// ===== op=get:查询异步结果(单次拉取,不内部轮询)=====
|
||||
async function queryAsyncResult({ token, id }) {
|
||||
const data = await callZhipu({
|
||||
return await callZhipu({
|
||||
url: ZHIPU_QUERY_URL(id),
|
||||
token,
|
||||
method: "GET",
|
||||
});
|
||||
return data; // { id, request_id, model, task_status, video_result?, ... }
|
||||
// { id, request_id, model, task_status, video_result?, ... }
|
||||
}
|
||||
|
||||
// ---------- op=create 的业务处理 ----------
|
||||
async function handleCreate({ args, headers, token }) {
|
||||
if (!args.prompt && !args.image_url) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: 'op=create requires at least one of "prompt" or "image_url".',
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
|
||||
// AI 不可见:全部走 header
|
||||
const model = pickModel(headers) || DEFAULT_MODEL;
|
||||
const quality = pickHeaderString(
|
||||
headers,
|
||||
"x-video-task-quality",
|
||||
DEFAULT_QUALITY
|
||||
);
|
||||
const fps = pickHeaderInt(headers, "x-video-task-fps", DEFAULT_FPS);
|
||||
const watermark_enabled = pickHeaderBool(
|
||||
headers,
|
||||
"x-video-task-watermark",
|
||||
DEFAULT_WATERMARK_ENABLED
|
||||
);
|
||||
|
||||
try {
|
||||
const data = await submitVideo({
|
||||
token,
|
||||
model,
|
||||
prompt: args.prompt,
|
||||
image_url: args.image_url,
|
||||
quality,
|
||||
with_audio: args.with_audio ?? false,
|
||||
fps,
|
||||
watermark_enabled,
|
||||
});
|
||||
|
||||
const summary =
|
||||
`Model: ${model}\n` +
|
||||
`task_id: ${data.id}\n` +
|
||||
`request_id: ${data.request_id}\n` +
|
||||
`task_status: ${data.task_status}\n` +
|
||||
`\n` +
|
||||
`Video generation is asynchronous. ` +
|
||||
`Call this tool again with op="get" and task_id="${data.id}" to fetch the result.`;
|
||||
|
||||
return { content: [{ type: "text", text: summary }] };
|
||||
} catch (e) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Failed to submit video task: ${e.message}${
|
||||
e.detail ? `\nDetails: ${JSON.stringify(e.detail)}` : ""
|
||||
}`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
}
|
||||
|
||||
// ---------- op=get 的业务处理 ----------
|
||||
async function handleGet({ args, token }) {
|
||||
if (!args.task_id) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [{ type: "text", text: 'op=get requires "task_id".' }],
|
||||
};
|
||||
}
|
||||
|
||||
try {
|
||||
const data = await queryAsyncResult({ token, id: args.task_id });
|
||||
const status = data.task_status || "UNKNOWN";
|
||||
|
||||
if (status === "SUCCESS") {
|
||||
const v = (data.video_result || [])[0];
|
||||
if (!v || !v.url) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Task succeeded but no video URL returned. Raw: ${JSON.stringify(
|
||||
data
|
||||
)}`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
const summary =
|
||||
`task_id: ${data.id}\n` +
|
||||
`model: ${data.model}\n` +
|
||||
`task_status: SUCCESS\n` +
|
||||
`Video URL: ${v.url}\n` +
|
||||
(v.cover_image_url ? `Cover: ${v.cover_image_url}\n` : "");
|
||||
return {
|
||||
content: [
|
||||
{ type: "text", text: summary },
|
||||
{ type: "text", text: v.url },
|
||||
],
|
||||
};
|
||||
}
|
||||
|
||||
// PROCESSING / FAIL / 其他:原样返回关键字段
|
||||
const errBlock = data.error ? `\nerror: ${JSON.stringify(data.error)}` : "";
|
||||
return {
|
||||
isError: status === "FAIL",
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text:
|
||||
`task_id: ${data.id}\n` +
|
||||
`model: ${data.model}\n` +
|
||||
`task_status: ${status}` +
|
||||
errBlock,
|
||||
},
|
||||
],
|
||||
};
|
||||
} catch (e) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Failed to query video task: ${e.message}${
|
||||
e.detail ? `\nDetails: ${JSON.stringify(e.detail)}` : ""
|
||||
}`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
}
|
||||
|
||||
function buildServer() {
|
||||
const server = new McpServer(
|
||||
{
|
||||
name: "video_gen",
|
||||
version: "1.0.0",
|
||||
},
|
||||
{
|
||||
capabilities: {
|
||||
tools: {},
|
||||
},
|
||||
}
|
||||
{ name: "video_task", version: "1.0.0" },
|
||||
{ capabilities: { tools: {} } }
|
||||
);
|
||||
|
||||
// ---------- 工具 1:video_gen ----------
|
||||
server.tool(
|
||||
"video_gen",
|
||||
`Submit an async video generation task. Returns a task_id; poll with query_video.
|
||||
"video_task",
|
||||
`Manages asynchronous video generation tasks.
|
||||
|
||||
- text-to-video: provide prompt only.
|
||||
- image-to-video: also provide image_url.`,
|
||||
op=create Submit a new task. Returns a task_id immediately; generation runs asynchronously.
|
||||
Requires at least one of prompt or image_url.
|
||||
- text-to-video: pass prompt only.
|
||||
- image-to-video: also pass image_url.
|
||||
op=get Fetch the status of a previously submitted task. Requires task_id.
|
||||
On SUCCESS, returns the video URL.
|
||||
|
||||
Typical flow: call op=create, then call op=get with the returned task_id until task_status is SUCCESS or FAIL.`,
|
||||
{
|
||||
op: z.enum(["create", "get"]).describe("Operation to perform."),
|
||||
prompt: z
|
||||
.string()
|
||||
.max(512)
|
||||
.optional()
|
||||
.describe(
|
||||
"Video text description (max 512 chars). Required; when combined with image_url it serves as the motion/action description."
|
||||
"Video text description (max 512 chars). Used with op=create. When combined with image_url it serves as the motion/action description."
|
||||
),
|
||||
image_url: z
|
||||
.string()
|
||||
.url()
|
||||
.optional()
|
||||
.describe(
|
||||
"Reference image for image-to-video. Must be a publicly reachable URL to a png/jpeg/jpg <= 5MB (for local files, first use the agent's built-in tool to mint a public temporary URL)."
|
||||
"Reference image for image-to-video. Used with op=create. Must be a publicly reachable URL to a png/jpeg/jpg <= 5MB (for local files, first use the agent's built-in tool to mint a public temporary URL)."
|
||||
),
|
||||
with_audio: z
|
||||
.boolean()
|
||||
.optional()
|
||||
.describe("Whether to generate AI audio. Default false."),
|
||||
},
|
||||
async (args, extra) => {
|
||||
const headers = extra?.requestInfo?.headers;
|
||||
const token = pickToken(headers);
|
||||
if (!token) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [{ type: "text", text: "Missing authentication token." }],
|
||||
};
|
||||
}
|
||||
|
||||
// AI 不可见:全部走 header
|
||||
const model = pickModel(headers) || DEFAULT_MODEL;
|
||||
const quality = pickHeaderString(headers, "x-video-gen-quality", DEFAULT_QUALITY);
|
||||
const fps = pickHeaderInt(headers, "x-video-gen-fps", DEFAULT_FPS);
|
||||
const watermark_enabled = pickHeaderBool(
|
||||
headers,
|
||||
"x-video-gen-watermark",
|
||||
DEFAULT_WATERMARK_ENABLED
|
||||
);
|
||||
|
||||
if (!args.prompt && !args.image_url) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: "Either prompt or image_url must be provided.",
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
|
||||
try {
|
||||
const data = await submitVideo({
|
||||
token,
|
||||
model,
|
||||
prompt: args.prompt,
|
||||
image_url: args.image_url,
|
||||
quality,
|
||||
with_audio: args.with_audio ?? false,
|
||||
fps,
|
||||
watermark_enabled,
|
||||
});
|
||||
|
||||
const summary =
|
||||
`Model: ${model}\n` +
|
||||
`task_id: ${data.id}\n` +
|
||||
`request_id: ${data.request_id}\n` +
|
||||
`task_status: ${data.task_status}\n` +
|
||||
`\n` +
|
||||
`Video generation is asynchronous. Call query_video with task_id="${data.id}" to fetch the result.`;
|
||||
|
||||
return {
|
||||
content: [{ type: "text", text: summary }],
|
||||
};
|
||||
} catch (e) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Failed to submit video task: ${e.message}${
|
||||
e.detail ? `\nDetails: ${JSON.stringify(e.detail)}` : ""
|
||||
}`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
}
|
||||
);
|
||||
|
||||
// ---------- 工具 2:query_video ----------
|
||||
server.tool(
|
||||
"query_video",
|
||||
"Query the status of a previously submitted video generation task. " +
|
||||
"Returns task_status and, if SUCCESS, the video URL(s).",
|
||||
{
|
||||
.describe(
|
||||
"Whether to generate AI audio. Used with op=create. Default false."
|
||||
),
|
||||
task_id: z
|
||||
.string()
|
||||
.min(1)
|
||||
.describe("task_id returned by video_gen."),
|
||||
.optional()
|
||||
.describe("task_id returned by op=create. Required when op=get."),
|
||||
},
|
||||
async (args, extra) => {
|
||||
const headers = extra?.requestInfo?.headers;
|
||||
@@ -312,69 +371,21 @@ function buildServer() {
|
||||
};
|
||||
}
|
||||
|
||||
try {
|
||||
const data = await queryAsyncResult({
|
||||
token,
|
||||
id: args.task_id,
|
||||
});
|
||||
|
||||
const status = data.task_status || "UNKNOWN";
|
||||
if (status === "SUCCESS") {
|
||||
const v = (data.video_result || [])[0];
|
||||
if (!v || !v.url) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Task succeeded but no video URL returned. Raw: ${JSON.stringify(data)}`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
const summary =
|
||||
`task_id: ${data.id}\n` +
|
||||
`model: ${data.model}\n` +
|
||||
`task_status: SUCCESS\n` +
|
||||
`Video URL: ${v.url}\n` +
|
||||
(v.cover_image_url ? `Cover: ${v.cover_image_url}\n` : "");
|
||||
switch (args.op) {
|
||||
case "create":
|
||||
return await handleCreate({ args, headers, token });
|
||||
case "get":
|
||||
return await handleGet({ args, token });
|
||||
default:
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{ type: "text", text: summary },
|
||||
{ type: "text", text: v.url },
|
||||
{
|
||||
type: "text",
|
||||
text: `Unknown op "${args.op}" (available: create, get).`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
|
||||
// PROCESSING / FAIL / 其他:原样返回关键字段
|
||||
const errBlock = data.error
|
||||
? `\nerror: ${JSON.stringify(data.error)}`
|
||||
: "";
|
||||
return {
|
||||
isError: status === "FAIL",
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text:
|
||||
`task_id: ${data.id}\n` +
|
||||
`model: ${data.model}\n` +
|
||||
`task_status: ${status}` +
|
||||
errBlock,
|
||||
},
|
||||
],
|
||||
};
|
||||
} catch (e) {
|
||||
return {
|
||||
isError: true,
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Failed to query video task: ${e.message}${
|
||||
e.detail ? `\nDetails: ${JSON.stringify(e.detail)}` : ""
|
||||
}`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
}
|
||||
);
|
||||
@@ -404,7 +415,7 @@ async function startHttp({ port, host }) {
|
||||
};
|
||||
|
||||
app.get("/health", (_req, res) => {
|
||||
res.json({ status: "ok", service: "video_gen", sessions: sessions.size });
|
||||
res.json({ status: "ok", service: "video_task", sessions: sessions.size });
|
||||
});
|
||||
|
||||
async function handleSessionRequest(req, res) {
|
||||
@@ -423,7 +434,8 @@ async function startHttp({ port, host }) {
|
||||
jsonrpc: "2.0",
|
||||
error: {
|
||||
code: -32000,
|
||||
message: "Bad Request: missing valid Mcp-Session-Id for non-initialize request",
|
||||
message:
|
||||
"Bad Request: missing valid Mcp-Session-Id for non-initialize request",
|
||||
},
|
||||
id: req.body?.id ?? null,
|
||||
});
|
||||
@@ -467,7 +479,7 @@ async function startHttp({ port, host }) {
|
||||
|
||||
httpServer = app.listen(port, host, () => {
|
||||
console.log(
|
||||
`[video_gen] MCP HTTP server listening on http://${host}:${port}/mcp`
|
||||
`[video] MCP HTTP server listening on http://${host}:${port}/mcp`
|
||||
);
|
||||
});
|
||||
}
|
||||
@@ -479,18 +491,18 @@ async function startStdio() {
|
||||
);
|
||||
const transport = new StdioServerTransport();
|
||||
await server.connect(transport);
|
||||
console.error("[video_gen] MCP stdio server started");
|
||||
console.error("[video] MCP stdio server started");
|
||||
}
|
||||
|
||||
const args = parseArgs(process.argv.slice(2));
|
||||
if (args.http) {
|
||||
startHttp({ port: args.port, host: args.host }).catch((e) => {
|
||||
console.error("[video_gen] failed to start http server:", e);
|
||||
console.error("[video] failed to start http server:", e);
|
||||
process.exit(1);
|
||||
});
|
||||
} else {
|
||||
startStdio().catch((e) => {
|
||||
console.error("[video_gen] failed to start stdio server:", e);
|
||||
console.error("[video] failed to start stdio server:", e);
|
||||
process.exit(1);
|
||||
});
|
||||
}
|
||||
Reference in New Issue
Block a user