SKP mcp
SkinKnowledgeBase provides source-backed skincare question answering and product comparison context for AI assistants through a public, read-only MCP endpoint.
README
SkinKnowledgeBase Skincare Question and Answer MCP
SkinKnowledgeBase MCP is a public, unauthenticated, read-only remote MCP server for source-backed skincare question-answering and skincare product comparison context.
- Website: https://skinknowledgebase.com
- MCP endpoint: https://mcp.skinknowledgebase.com/mcp
- AI access docs: https://skinknowledgebase.com/ai-access
- Developer docs: https://skinknowledgebase.com/developers
- Server card: https://mcp.skinknowledgebase.com/.well-known/mcp/server-card.json
- Privacy: https://skinknowledgebase.com/privacy
- Terms: https://skinknowledgebase.com/terms
- Contact: https://skinknowledgebase.com/contact
What it does
The SkinKnowledgeBase Skincare Question and Answer MCP exposes public SkinKnowledgeBase content through read-only MCP retrieval tools. It helps AI assistants answer cosmetic-skincare and skincare-product questions with cited context for:
- skincare questions
- ingredients
- products
- skin concerns
- side effects
- cited sources
- public product comparison pages
It is designed for source-backed skincare answers, transparent unknowns, and canonical citations — not unsupported beauty claims.
Transport
- Type: remote HTTPS MCP
- Endpoint:
https://mcp.skinknowledgebase.com/mcp - Auth: none
- Mutations/destructive actions: none
- Medical advice: no
Clients may describe this as Streamable HTTP or SSE-compatible remote MCP depending on their field labels.
Tools
The live server currently exposes read-only retrieval tools including:
search— compact ranked SKB search across supported entity/resource types.fetch— fetch one canonical SKB entity or public-safe question/product/comparison-page resource.search_questions— search published MCP-eligible Question entities.get_question_bundle— fetch a published MCP-eligible Question bundle.get_entity— fetch one published MCP-eligible SKB entity by type and slug/ID.search_entities— search published MCP-eligible SKB entities.get_ranked_products_for_question— supporting product suggestions for an eligible skincare Question, capped at 2.get_ranked_product_for_question— deprecated/backcompat alias for supporting product suggestions.get_sources_for_question— cited Source entities for a Question.list_entities— list published MCP-eligible SKB entities, optionally filtered by entity type.
Example use cases
- Answer skincare and skincare-product questions with citations and canonical SKB URLs.
- Retrieve ingredient, product, concern, side-effect, and source context for an AI assistant.
- Support product comparisons using public product facts, ingredient signals, source confidence, and caveats.
- Surface what is known, unknown, and source-backed.
Appropriate use and medical boundary
SkinKnowledgeBase is educational cosmetic-skincare reference content. It is not medical advice, diagnosis, or treatment. It does not provide guaranteed outcomes, paid rankings, fake ratings, or “best overall” clinical claims. Medical symptoms, prescriptions, procedures, or urgent-care decisions should be handled by qualified clinicians.
Client configuration
See docs/llms-install.md for copy/paste configuration examples.
Registry metadata
- Official display name:
SkinKnowledgeBase Skincare Question and Answer MCP - Tagline:
Skincare Question and Answer and Product Comparison for AI Assistants and Search - Tags:
mcp,remote-mcp,streamable-http,skincare,question-answering,citations,ingredients,products,product-comparison,read-only
License
Documentation in this repository is released under the MIT License. SkinKnowledgeBase website content remains subject to the website terms.
Glama / local scanner bridge
This repository includes a minimal Dockerfile for MCP registries that require a runnable container. It starts mcp-remote and bridges local MCP stdio introspection to the hosted SkinKnowledgeBase endpoint:
docker build -t skb-mcp .
docker run --rm -i skb-mcp
The source of truth remains the hosted read-only MCP endpoint: https://mcp.skinknowledgebase.com/mcp.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。