Skip to main content
Glama
ah-wq

MCP Vision Relay

by ah-wq
OverviewSchemaRelated ServersScoreDiscussions
TypeScript
Local

MCP Vision Relay

English | 中文说明

MCP Vision Relay 将本地安装的多模态 CLI（目前支持 Google Gemini CLI 与 Qwen CLI）封装为 Model Context Protocol (MCP) 服务器，帮助 Claude、Codex 等支持 MCP 的工具直接使用它们的看图能力。

Why it matters: when Claude Code connects to providers such as k2, DeepSeek, or MiniMax M2, the backing models today are text-only—no built-in vision. By relaying calls through Gemini/Qwen CLI, MCP Vision Relay gives those deployments an inexpensive path to regain multimodal features without switching providers.

Features

  • Unified image tools – gemini_analyze_image 与 qwen_analyze_image 两个工具覆盖路径、URL、base64 三种输入。

  • Provider relay architecture – 通过 provider 抽象切换或扩展不同 CLI，保留模型、输出格式等配置能力。

  • Robust input handling – 自动校验图片大小、扩展名，必要时下载或写入临时文件并在使用后清理。

  • Configurable execution – 支持可选 sandbox、超时、额外旗标、模型覆盖，以及 .env/环境变量配置。

  • Actionable outputs – 对 stdout 进行规整并附加元信息，便于客户端在 UI 中展示或后续处理。

Related MCP server: 2Xapi.com GPT-image MCP Server

Quick Start

Prerequisites

  1. Node.js ≥ 18

  2. 已安装并能在命令行直接调用的 Gemini CLI 与/或 Qwen CLI

  3. 对应 CLI 的登陆/鉴权已完成（确保 gemini -p "hi"、qwen -p "hi" 能返回结果）

Install & Build

npm install
npm run build

Run the MCP server

# 开发模式（tsx 直接启动 TypeScript）
npm run dev

# 生产模式（使用编译产物）
npm run start

服务器通过 stdio 与 MCP 客户端通信，适用于 claude mcp add、codex mcp add 等命令。

MCP Integration Examples

⚠️ When registering the server with an MCP client, invoke the entry point directly. Running npm run dev inside the registration command causes npm to print a banner on stdout, which breaks the MCP handshake. If you must keep the npm script, wrap it with npm --silent run dev -- so no extra text reaches stdout.

Claude

claude mcp add mcp-vision-relay -- npx tsx /absolute/path/to/mcp-vision-relay/src/index.ts

Codex CLI

codex mcp add mcp-vision-relay -- node /absolute/path/to/mcp-vision-relay/dist/index.js

完成后即可在会话/任务中选择 mcp-vision-relay 的工具调用。

Available Tools

gemini_analyze_image

分析图像并返回 Gemini CLI 给出的描述。

参数

类型

说明

image

string

必填；本地路径、HTTP(S) URL 或 base64 字符串

prompt

string?

额外指令，默认使用 GEMINI_DEFAULT_PROMPT

model

string?

覆盖默认模型（如 gemini-2.0-flash）

outputFormat

"text" | "json"?

控制 -o 输出格式

sandbox

boolean?

是否添加 -s sandbox 旗标

extraFlags

string[]?

附加自定义参数

timeoutMs

number?

CLI 超时（默认 120000ms）

返回内容包含一段 Markdown 文本与元信息（模型、输入来源、耗时等）。

qwen_analyze_image

走 Qwen CLI 进行图像理解。逻辑与 Gemini 类似，但会在需要时自动把本地文件转成 data URL 供 CLI 读取。

参数

类型

说明

image

string

必填；本地路径、HTTP(S) URL 或 base64 字符串

prompt

string?

额外指令，默认使用 QWEN_DEFAULT_PROMPT

model

string?

Qwen 模型（如 qwen2.5-omni-medium）

sandbox

boolean?

是否加 -s

extraFlags

string[]?

附加参数

timeoutMs

number?

CLI 超时（默认 120000ms）

Configuration

复制 .env.example 进行配置，常用条目：

  • GEMINI_CLI_COMMAND / QWEN_CLI_COMMAND：CLI 可执行文件路径。

  • GEMINI_DEFAULT_MODEL / QWEN_DEFAULT_MODEL：默认模型名。

  • GEMINI_OUTPUT_FORMAT：控制 Gemini 输出（text 或 json）。

  • MCP_COMMAND_TIMEOUT_MS：全局超时（毫秒）。

  • MCP_MAX_IMAGE_BYTES：允许的最大图片大小。

  • MCP_ALLOWED_IMAGE_EXTENSIONS：允许的扩展名列表。

  • MCP_IMAGE_TEMP_DIR：存放下载/解码后临时文件的目录。

如需针对单次调用覆盖 CLI 命令，可在工具参数中提供 cliPath（Gemini/Qwen 均支持）。

Local Diagnostics

项目提供两个简单脚本，便于手动验证 CLI 调用：

npx tsx scripts/dev/manual-gemini-test.ts
npx tsx scripts/dev/manual-qwen-test.ts

确保在执行前已完成 npm run build 或使用 ts-node/tsx。

Project Structure

src/
  index.ts                 # 程序入口，加载 env 并启动 MCP 服务器
  config/                  # 配置解析（appConfig 等）
  providers/               # CLI provider 适配层（Gemini/Qwen）
  server/                  # MCP server wiring
  tools/                   # MCP 工具定义与注册
  utils/                   # 文件、CLI 调度等公共工具
scripts/
  dev/                     # 手动验证脚本
test-assets/               # 示例图像资源

Roadmap

  • ✅ Gemini CLI 图像分析

  • ✅ Qwen CLI 图像分析（含自动 data URL）

  • ⏳ 资源列表与更多多模态 provider

  • ⏳ 自动化测试与 lint/format pipeline

License

MIT License. 欢迎 issue / PR 贡献改进。

Acknowledgements

Available Tools

2 tools
gemini_analyze_imageC

Use Google Gemini CLI to describe or analyze an image using multimodal capabilities.

ParametersJSON Schema
NameRequiredDescriptionDefault
imageYesLocal file path, http(s) URL, or base64-encoded image to analyze.
promptNoInstruction for Gemini. Defaults to config value.
modelNoGemini model identifier (e.g., gemini-2.0-flash).
cliPathNoOverride the Gemini CLI executable path.
sandboxNoWhether to run the CLI with the sandbox flag (-s).
outputFormatNoRequest Gemini CLI to return the specified output format.
extraFlagsNoAdditional CLI flags to append as-is.
timeoutMsNoMaximum time (in milliseconds) to wait for CLI execution.

TDQS

C2.9/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries full burden for behavioral disclosure but offers minimal information. It mentions 'multimodal capabilities' but doesn't explain what this means operationally, nor does it cover important behavioral aspects like rate limits, authentication requirements, error handling, or what happens when the CLI fails. The description is too vague about the tool's actual behavior.

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?

The description is extremely concise at just one sentence with zero wasted words. It's front-loaded with the core functionality and efficiently communicates the essential purpose without unnecessary elaboration. Every word earns its place in this minimal description.

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?

For a tool with 8 parameters, no annotations, and no output schema, the description is inadequate. It doesn't explain what the tool returns, how to interpret results, error conditions, or operational constraints. The description fails to compensate for the lack of structured metadata, leaving significant gaps in understanding how to effectively use this complex tool.

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?

With 100% schema description coverage, the schema already documents all 8 parameters thoroughly. The description adds no additional parameter semantics beyond what's in the schema, so it meets the baseline of 3. The description doesn't explain parameter interactions, default behaviors, or provide examples of effective parameter combinations.

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 action ('describe or analyze an image') and the technology used ('Google Gemini CLI with multimodal capabilities'), which provides a specific verb+resource combination. However, it doesn't explicitly differentiate from its sibling 'qwen_analyze_image' beyond mentioning the different technology stack.

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?

The description provides no guidance on when to use this tool versus alternatives, including its sibling 'qwen_analyze_image'. There's no mention of specific use cases, prerequisites, or comparative advantages that would help an agent choose between available image analysis tools.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

qwen_analyze_imageC

Use Qwen CLI to describe or analyze an image with its multimodal capabilities.

ParametersJSON Schema
NameRequiredDescriptionDefault
imageYesLocal file path, http(s) URL, or base64-encoded image to analyze.
promptNoInstruction for Qwen. Defaults to config value.
modelNoQwen model identifier.
cliPathNoOverride the Qwen CLI executable path.
sandboxNoWhether to run the CLI with the sandbox flag (-s).
extraFlagsNoAdditional CLI flags to append as-is.
timeoutMsNoMaximum time (in milliseconds) to wait for CLI execution.

TDQS

C2.9/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden of behavioral disclosure. The description mentions 'multimodal capabilities' but doesn't explain what this entails (e.g., types of analysis, output format, limitations). It also lacks details on permissions, rate limits, error handling, or what happens during execution (e.g., whether it's synchronous). This leaves significant gaps for an agent to understand the tool's behavior.

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?

The description is a single, efficient sentence that directly states the tool's purpose without unnecessary words. It's appropriately sized and front-loaded, with every element ('Use Qwen CLI', 'describe or analyze an image', 'multimodal capabilities') contributing essential information. There's zero waste in the phrasing.

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?

Given the complexity (7 parameters, no annotations, no output schema), the description is insufficiently complete. It doesn't explain the return values or output format, which is critical since there's no output schema. It also lacks behavioral context (e.g., what 'analyze' entails, error cases, or performance characteristics). For a tool with this many parameters and no structured support, the description should provide more guidance.

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 description coverage is 100%, so the input schema fully documents all 7 parameters with clear descriptions. The tool description adds no additional parameter information beyond what's in the schema. According to the rules, when schema coverage is high (>80%), the baseline score is 3 even with no param info in the description, which applies here.

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: 'describe or analyze an image with its multimodal capabilities' using Qwen CLI. It specifies the verb ('describe or analyze'), resource ('image'), and technology ('Qwen CLI'), making the purpose unambiguous. However, it doesn't explicitly differentiate from its sibling 'gemini_analyze_image' beyond mentioning Qwen specifically.

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?

The description provides no guidance on when to use this tool versus alternatives. It doesn't mention the sibling tool 'gemini_analyze_image' or any other alternatives, nor does it provide context about when Qwen might be preferred over other image analysis tools. Usage is implied through the tool name and description but not explicitly stated.

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. Dates show when Glama detected each change.

  1. 2 tool updates
    • First observedgemini_analyze_image
    • First observedqwen_analyze_image
Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

C2.8/5.0
Disambiguation1/5

The two tools are essentially identical in purpose—both describe or analyze images using multimodal capabilities, differing only in the underlying model (Gemini vs. Qwen). An agent would have no clear basis to choose one over the other based on their descriptions, leading to confusion and misselection.

Naming Consistency5/5

The tool names follow a perfectly consistent pattern: both use a clear 'model_verb_noun' structure (gemini_analyze_image, qwen_analyze_image). This consistency makes it easy to understand what each tool does at a glance.

Tool Count2/5

With only two tools, the server feels thin for a vision-related domain, as it lacks coverage for common operations like image generation, editing, or filtering. The tools are redundant in functionality, making the count seem artificially low for the apparent scope.

Completeness2/5

The tool surface is severely incomplete for a vision server; it only offers image analysis via two similar models, with no support for tasks like image creation, transformation, or retrieval. This creates significant gaps that will limit agent capabilities in handling broader vision workflows.

Maintenance

ActivityInactive
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

Related MCP Servers

Appeared in Searches

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/ah-wq/mcp-vision-relay'

If you have feedback or need assistance with the MCP directory API, please join our Discord server