Skip to main content
Glama
Sign In
mclans-image-mcp

mclans-image-mcp

by YupegLV
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Local

mclans-image-mcp

异步生图 MCP Server — 提交即忘，图片自动下载到本地。

配置 MCP

Claude Code

编辑 ~/.claude/settings.json：

{
  "mcpServers": {
    "mclans-image": {
      "command": "npx",
      "args": ["-y", "github:YupegLV/mclans-create-image-mcp"],
      "env": {
        "IMAGE_API_KEY": "sk-xxxxx"
      }
    }
  }
}

Claude Desktop

编辑配置文件：

  • Windows: %APPDATA%\Claude\claude_desktop_config.json

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "mclans-image": {
      "command": "npx",
      "args": ["-y", "github:YupegLV/mclans-create-image-mcp"],
      "env": {
        "IMAGE_API_KEY": "sk-xxxxx"
      }
    }
  }
}

Cursor / VS Code

在设置中搜索 MCP，添加服务器配置：

{
  "mcp": {
    "servers": {
      "mclans-image": {
        "command": "npx",
        "args": ["-y", "github:YupegLV/mclans-create-image-mcp"],
        "env": {
          "IMAGE_API_KEY": "sk-xxxxx"
        }
      }
    }
  }
}

API Key 配置说明

方式

说明

环境变量 IMAGE_API_KEY

推荐，在 MCP 配置的 env 中传入

命令行参数

在 args 中添加 --api-key=sk-xxxxx

配置文件

手动创建 ~/.mclans-image-mcp/config.json

首次通过环境变量或命令行参数启动后，配置会自动持久化到 ~/.mclans-image-mcp/config.json，后续无需重复配置。

如需自定义 API 地址，可添加环境变量 IMAGE_API_BASE_URL 或命令行参数 --api-base-url=https://your-api-url。

Available Tools

5 tools
check_taskA

查询生图任务的状态和结果。后台 Worker 会自动轮询远程服务器并下载图片，无需手动轮询。任务完成后返回本地图片路径

Inputs: taskId (required).

generate_imageA

提交异步生图任务，后台 Worker 会自动轮询远程服务器并下载图片到本地。图片保存到 projectDir/outputSubdir/ 目录下（默认 mclans-image 子文件夹，可通过 outputSubdir 参数自定义）。任务提交后立即返回，生图通常需要 5-10 分钟。请询问用户：是需要我持续跟踪任务进度直到完成，还是提交后由系统自动处理、后续按需查询？

Inputs: prompt (required), projectDir (required), fileName (required), count, +1 more.

generate_turnaroundA

提交异步角色多视图生成任务，根据参考图片生成 2x2 网格（正面/侧面/背面/动作），保持角色一致性。图片保存到 projectDir/outputSubdir/ 目录下（默认 mclans-image 子文件夹，可通过 outputSubdir 参数自定义）。请询问用户：是需要我持续跟踪任务进度直到完成，还是提交后由系统自动处理、后续按需查询？

Inputs: prompt (required), imagePath (required), projectDir (required), fileName (required), +1 more.

generate_videoA

提交异步视频生成任务，后台 Worker 会自动轮询远程服务器并下载视频到本地。视频保存到 projectDir/outputSubdir/ 目录下（默认 mclans-image 子文件夹，可通过 outputSubdir 参数自定义）。支持文生视频和图生视频。任务提交后立即返回，视频生成通常需要较长时间。请询问用户：是需要我持续跟踪任务进度直到完成，还是提交后由系统自动处理、后续按需查询？

Inputs: prompt (required), projectDir (required), fileName (required), model, +4 more.

list_tasksC

列出所有生图任务，可按状态筛选

Inputs: status.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A3.7/5.0

    Scored across 5 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: generating images, generating turnaround (multi-view), generating videos, checking task status, and listing tasks. There is no overlap or ambiguity.

    Naming Consistency4/5

    Tool names follow a consistent verb_noun pattern in snake_case (e.g., generate_image, check_task, list_tasks). The only minor deviation is 'generate_turnaround' where 'turnaround' is a specific noun, but still fits the pattern.

    Tool Count5/5

    5 tools is well-scoped for an image and video generation server. It covers submission (3 generation types), status checking, and listing, without being overwhelming or too sparse.

    Completeness4/5

    The tool surface covers core workflows: submit generation tasks (image, turnaround, video), check results, and list tasks. However, there are no tools for cancelling or deleting tasks, which are minor gaps.

    Resources

    Related MCP Connectors

    Related MCP Servers