{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_j84m3semf4yc","handle":"mcp-image-cutout","url":"https://wellknown.network/agents/mcp-image-cutout","links":{"self":"https://wellknown.network/agents/mcp-image-cutout/record.json","html":"https://wellknown.network/agents/mcp-image-cutout","markdown":"https://wellknown.network/agents/mcp-image-cutout/record.md","api":"https://wellknown.network/api/v1/agents/mcp-image-cutout","status":"https://wellknown.network/api/v1/agents/mcp-image-cutout/status","claim":"https://wellknown.network/agents/mcp-image-cutout/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/mcp-image-cutout/claim.json","badge":"https://wellknown.network/agents/mcp-image-cutout/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/mcp-image-cutout/history","tools":"https://wellknown.network/api/v1/agents/mcp-image-cutout/tools"},"ard":{"identifier":"urn:air::server:mcp-image-cutout","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"mcp-image-cutout","summary":"MCP服务器：图像编辑工具，提供显著性分割、背景移除等图像处理功能","description":"# MCP 图像抠图服务器\n\n这是一个基于Model Context Protocol (MCP)的服务器，专门提供火山引擎图像抠图功能。\n\n## 功能特性\n\n- **智能抠图**: 使用火山引擎的显著性分割技术，自动识别并抠出图像中的主要对象\n- **批量处理**: 支持同时处理多张图像\n- **自动上传**: 抠图结果自动上传到服务器并返回URL链接\n- **本地保存**: 可选择保存抠图结果到本地文件\n- **高精度**: 基于显著性检测的精确分割算法\n\n## 安装\n\n### 从 PyPI 安装\n\n```bash\npip install mcp-image-cutout\n```\n\n### 从源码安装\n\n```bash\ngit clone https://github.com/fengjinchao/mcp-image-cutout.git\ncd mcp-image-cutout\npip install -e .\n```\n\n### 2. 配置API密钥\n\n推荐使用环境变量设置API密钥：\n\n```bash\nexport VOLC_ACCESS_KEY=\"your_access_key\"\nexport VOLC_SECRET_KEY=\"your_secret_key\"\n```\n\n或者在代码中直接设置（不推荐用于生产环境）。\n\n### 3. 运行服务器\n\n```bash\n# 使用命令行工具\nmcp-image-cutout\n\n# 或直接运行模块\npython -m mcp_image_cutout.server\n```\n\n## 在Claude Desktop中使用\n\n在Claude Desktop的配置文件中添加以下配置：\n\n**macOS/Linux**: `~/Library/Application Support/Claude/claude_desktop_config.json`\n\n```json\n{\n  \"mcpServers\": {\n    \"抠图工具\": {\n      \"command\": \"mcp-image-cutout\",\n      \"env\": {\n        \"VOLC_ACCESS_KEY\": \"your_access_key\",\n        \"VOLC_SECRET_KEY\": \"your_secret_key\"\n      }\n    }\n  }\n}\n```\n\n## 可用工具\n\n### image_cutout\n智能图像抠图，使用显著性分割自动识别并抠出图像中的主要对象，自动上传到服务器并返回图片URL。\n\n**参数**:\n- `image_urls`: 图像URL列表，支持多张图像同时处理\n\n**返回**:\n- 单张图片：直接返回图片URL\n- 多张图片：返回所有图片URL的列表\n\n**示例**:\n```\n请帮我抠出这张图片中的主要对象：https://example.com/image.jpg\n```\n\n```\n批量抠图这些图片：\n- https://example.com/image1.jpg\n- https://example.com/image2.jpg\n```\n\n## 抠图原理\n\n使用的 `saliency_seg` 显著性分割算法：\n- 基于视觉显著性检测图像中最重要的区域\n- 精确分割显著对象的轮廓\n- 生成高质量的抠图结果\n- 适用于各种复杂背景的图像处理\n\n## 上传功能\n\n抠图完成后，系统会自动：\n1. 将base64编码的图像数据转换为PNG文件\n2. 上传到指定服务器：`https://www.mcpcn.cc/api/fileUploadAndDownload/uploadMcpFile`\n3. 返回可访问的图片URL链接\n4. 支持批量上传多张抠图结果\n\n**上传接口返回格式**：\n```json\n{\n    \"code\": 0,\n    \"data\": {\n        \"url\": \"https://juezhi.oss-cn-shanghai.aliyuncs.com/file/uploads/mcp/xxx.webp\"\n    },\n    \"msg\": \"成功\"\n}\n```\n\n## 注意事项\n\n1. 确保图像URL可以公开访问\n2. 处理结果会以base64格式返回，大图像可能需要较长处理时间\n3. API调用有频率限制，请合理使用\n4. 生产环境中请使用环境变量设置API密钥\n\n## 故障排除\n\n### 常见问题\n\n1. **服务器无法启动**\n   - 检查Python版本（需要3.10+）\n   - 确认所有依赖已正确安装\n\n2. **API调用失败**\n   - 验证API密钥是否正确\n   - 检查网络连接\n   - 确认图像URL…","publisher":null,"homepage":"https://github.com/fengjinchao/mcp-image-cutout#readme","repository":"https://github.com/fengjinchao/mcp-image-cutout#readme","version":"0.2.9","license":"MIT","protocols":["mcp"],"tags":["ai","computer-vision","cutout","image","mcp","volcengine"],"pricing":null,"endpoints":[{"url":"pypi:mcp-image-cutout","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi"}},"derived":{"capabilities":[{"slug":"dev.version-control","name":"Version Control","confidence":0.745,"provenance":"derived"},{"slug":"media.image-understanding","name":"Image Understanding","confidence":0.638,"provenance":"derived"}],"categories":["dev","media"],"language":"fr"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"mcp-image-cutout","registry":"pypi","observedAt":"2026-10-22T09:23:41.732Z","publishedAt":"2026-04-25T09:00:09.004869Z","latestVersion":"0.2.9"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"mcp-image-cutout","url":"https://pypi.org/project/mcp-image-cutout/","firstSeenAt":"2026-09-10T02:22:50.320Z","fetchedAt":"2026-10-22T09:22:21.576Z","normalizedAt":"2026-10-22T09:22:21.576Z"}]},"firstSeenAt":"2026-09-10T02:22:50.320Z","updatedAt":"2026-10-22T09:23:41.732Z"}