A
licenseA
qualityD
maintenanceEnables image analysis using any OpenAI-compatible vision API, supporting URLs, local files, or base64 input with custom prompts.
1
MIT
Doubao Vision MCP Server
by kira4094
This server provides Doubao-powered multimodal capabilities via Volcengine Ark, using a single API key.
Image understanding (
doubao_vision_understand): Analyze local or URL images with a text prompt, adjustable detail (auto/low/high),max_tokens, andtemperature.Image generation (
doubao_seedream_generate): Generate images from text prompts or reference images, with configurable size, aspect ratio, and number of images.Video generation (
doubao_seedance_generate): Asynchronously submit text-to-video or image-to-video tasks; returns atask_idfor polling.Video status query (
doubao_seedance_query): Poll task status bytask_idand retrieve a download URL when succeeded.Server behavior can be configured through environment variables for vision, image, and video models, supporting preset or custom endpoints (
ep-xxxxx).
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Doubao Vision MCP Serverdescribe the objects and colors in this image: https://example.com/photo.jpg"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Doubao Vision MCP Server
火山方舟（Volcengine Ark）豆包系列 MCP Server：视觉理解 + Seedream 图像生成 + Seedance 视频生成，一套密钥全打通。
特性
🖼️ 4 个工具：视觉理解 / 图像生成 / 视频提交 / 视频轮询
🎬 Seedream 同步出图：图像几秒返回，无需轮询
🎥 Seedance 两段式：异步提交 task_id → 轮询拿视频（MCP stdio 不超时）
💬 自定义 prompt，支持中文/英文
🔄 所有模型可用环境变量切换，代码零改动
⚡ 一行 npx 部署
💰 支持火山方舟赠送的免费 token
Related MCP server: Vision MCP Server
支持的模型
视觉理解
项目 | 说明 |
模型 ID |
|
接入点 | 预置推理接入点（无需创建，直接填模型名） |
模态 | 文本 + 图片 + 语音 + 视频 |
图像生成（Seedream）
项目 | 说明 |
模型 ID |
|
图生图 |
|
备选 |
|
视频生成（Seedance）
项目 | 说明 |
模型 ID |
|
备选 |
|
⚠️ 注意 | 需在火山方舟控制台开通对应模型； |
环境变量
变量 | 必填 | 默认值 | 说明 |
| ✅ | — | 火山方舟 API Key。获取：控制台 → API Key 管理 |
| 否 |
| 视觉模型名（预置接入点）或 ep-xxxxx |
| 否 |
| Seedream 文生图模型 |
| 否 |
| Seedream 图生图模型（传 |
| 否 |
| Seedance 视频生成模型 |
| 否 |
| API 地址，通常无需修改 |
开通模型（重要）
在 火山方舟控制台 → 开通管理 开通需要的模型（通常有免费试用额度）：
图像生成：
Doubao Seedream系列视频生成：
Doubao Seedance系列（建议 2.0 / 2.5）
快速开始
npx 部署
{
"mcpServers": {
"doubao-vision": {
"command": "npx",
"args": ["-y", "doubao-vision-mcp-server"],
"env": {
"ARK_API_KEY": "ark-xxxxxxxxx-xxxxx"
}
}
}
}自定义模型
{
"mcpServers": {
"doubao-vision": {
"command": "npx",
"args": ["-y", "doubao-vision-mcp-server"],
"env": {
"ARK_API_KEY": "ark-xxxxxxxxx-xxxxx",
"SEEDREAM_MODEL": "doubao-seedream-5-0-pro-260628",
"SEEDANCE_MODEL": "doubao-seedance-2-5-260628"
}
}
}
}工具
doubao_vision_understand — 视觉理解
参数 | 必填 | 默认值 | 说明 |
| ✅ | — | 本地图片路径 |
| ✅ | — | 对图片的指令，越具体越好 |
| 否 |
| 图片精度： |
| 否 |
| 最大输出 token 数 |
| 否 |
| 采样温度（0~2） |
doubao_seedream_generate — 图像生成（同步）
参数 | 必填 | 默认值 | 说明 |
| ✅ | — | 图像描述（中英文皆可） |
| 否 | — | 参考图（本地路径/URL），传了即图生图 |
| 否 | — | 分辨率，如 |
| 否 | — | 宽高比： |
| 否 |
| 生成数量 |
doubao_seedance_generate — 视频生成（异步提交）
参数 | 必填 | 默认值 | 说明 |
| ✅ | — | 视频描述 |
| 否 | — | 参考图（图生视频） |
| 否 | — | 分辨率（如 |
| 否 | — | 时长（秒，模型支持范围 4~30） |
返回 task_id，随后用 doubao_seedance_query 轮询。
doubao_seedance_query — 视频生成（轮询）
参数 | 必填 | 说明 |
| ✅ |
|
返回 status（queued / running / succeeded / failed），succeeded 时给出视频下载 URL。
使用示例
分析这张 UI 截图：
C:\screenshot.png，描述它的布局和配色方案
生成一张暖色调复古酒吧氛围图，两个人西装对话，电影胶片质感
用这张参考图生成一段 5 秒视频：角色转头看向镜头，微笑（会返回 task_id，再查询）
本地开发
git clone https://github.com/kira4094/doubao-vision-mcp-server.git
cd doubao-vision-mcp-server
npm install
node src/index.js验证结果
能力 | 模型 | 结果 |
视觉理解 |
| ✅ |
图像生成（文生图） |
| ✅ 实测出图 |
图像生成（图生图） |
| ✅ 模型已开通 |
视频生成 |
| ⏳ 需控制台开通后启用 |
关联项目
glm-vision-mcp-server — 智谱 GLM 视觉模型 MCP
agnes-image-mcp-server — Agnes 图像生成 MCP
agnes-video-mcp-server — Agnes 视频生成 MCP
Available Tools
1 tooldoubao_vision_understandC
Analyze an image using Doubao vision model via Volcengine Ark API. Supports both preset inference (model name) and custom inference (ep-xxxxx endpoint ID). Configure via DOUBAO_MODEL environment variable.
ParametersJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| image | Yes | Image source: local file path (e.g. C:/path/to/screenshot.png) or URL (https://...) | |
| prompt | Yes | What to ask about the image. Be specific for best results. | |
| detail | No | Image detail level. 'high' for fine-grained analysis | auto |
| max_tokens | No | Maximum output tokens | |
| temperature | No | Sampling temperature (0-2) |
TDQS
C2.9/5.0
Behavior2/5
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden. It does not disclose authentication needs, rate limits, failure behaviors (e.g., invalid image), or the response format. Minimal behavioral information is provided beyond the basic API call.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Conciseness5/5
Is the description appropriately sized, front-loaded, and free of redundancy?
Two concise sentences with no wasted words. Information is front-loaded and efficiently communicates the core functionality and configuration option.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Completeness2/5
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Missing output schema and behavioral details. The description does not explain what the tool returns or how to interpret results. For a vision analysis tool with multiple parameters, more context is needed for completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Parameters3/5
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so baseline is 3. The description adds little beyond the schema: it mentions image can be file path or URL and prompt specificity, but these are already in schema descriptions. The description does not explain how parameters interact or provide additional context.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Purpose4/5
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Analyze an image using Doubao vision model via Volcengine Ark API.' It mentions both preset and custom inference modes, providing good specificity. However, 'analyze' is somewhat broad, and without sibling tools, differentiation isn't needed.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Usage Guidelines2/5
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance on when to use this tool versus alternatives. It mentions configuration via environment variable as a prerequisite, but lacks context for optimal use cases or exclusions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
1 tool update
v1.0.0- First observed
doubao_vision_understand
TDQS
B3.2/5.0
Scored across 1 tool
Disambiguation5/5
Only one tool exists, so there is no ambiguity or overlap with other tools.
Naming Consistency5/5
With a single tool, naming consistency is not applicable; the name is clear and descriptive.
Tool Count3/5
One tool is on the low end of the range, but it is focused and covers the core image analysis task for a specialized vision server.
Completeness3/5
The tool covers image analysis, but the server lacks additional operations like image generation or batch processing that might be expected in a vision server.
Maintenance
ActivitySlowing
ResponsivenessNo issues
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
OCR, transcription, file extraction, and image generation for AI agents via MCP.
Generate AI images and videos from any compatible MCP client.
Create images & video from any MCP agent — 17 models, spend limits, one URL.
MCP server for Qwen Image 3 AI image generation
Related MCP Servers
- FlicenseNot gradedqualityBmaintenanceProvides image understanding capabilities for MCP clients (e.g., Claude Code) by analyzing images using vision models from providers like Alibaba Cloud Bailian, OpenAI, or OpenRouter, returning detailed descriptions in Markdown format.1-
- AlicenseNot gradedqualityCmaintenanceEnables analysis of local images through Kimi (Moonshot AI) vision models via the MCP protocol, supporting features like OCR and long context understanding.37MIT
- FlicenseNot gradedqualityBmaintenanceEnables image understanding and OCR through Xiaomi's MiMo vision language model, providing tools for image description, Q&A, and text recognition via MCP. Supports both image URLs and local file paths.-