Site Rag简化的RAG管道为提取和查询网站内容

4.3 (4)

审阅者 Daniel Nikulshyn·更新 2026年7月

RAG Developer Tools Web Scraping Vector Search LLM Knowledge Base Embeddings

概览

Site Rag 是一款检索增强生成（RAG）工具，旨在将网站内容转化为可搜索的知识库。它负责抓取、解析和 embedding 步骤，使网页可以被大型语言模型查询，从而减少构建自定义 RAG 系统所需的繁琐工作。该工具面向希望在文档网站、博客或其他公开网页资源上快速搭建问答层的开发者和团队。通过将提取和查询包装成统一的工作流，它让用户能够专注于提示和下游应用逻辑，而非基础设施。

主要功能

自动 web 内容提取
嵌入和向量存储
自然语言查询
RAG 管道orchestration
面向开发者的工作流

价格

模型: Free
分类: 系统机枢运器
评分: 4.3 / 5 (4)

使用场景

产品文档中的问答

爬取文档网站并以自然语言的问答接口形式显示它，允许用户询问问题并从文档中获得定基答案。

可搜索博客知识库

将公司的博客或内容存档转化为可检索的知识库，使得可以通过LLM驱动的搜索轻松检索相关帖子和见解。

内部支持助手

构建一个参考公共Web源以帮助支持团队快速查找准确答案而不是手动浏览浏览器的助手。

快速原型 RAG 应用

跳过从头搭建提取、嵌入和向量存储，从而允许开发者重点处理促进和应用逻辑以构建新的基于RAG的产品。

优点 & 缺点

优点

简化了end-to-end RAG 安装过程
目的构建于 web 内容
减少了开发者样板代码
适用于文档和知识库
用途

缺点

仅限于基于网站的源
需要技术设置
质量取决于站点结构

对决战绩

在万神殿中参与了 3 对决。

第1

第2

第3

Last 3 battles

评测

4.3

4 个评分的平均值。

登录以留下评测。

Camille Laurent

Apr 22, 2026

Solid for our team

We rolled this out across the team last quarter and purpose-built for web content. Embedding and vector storage fits neatly into how we already work, and embedding and vector storage removed a step we used to do by hand. Limited to website-based sources, which is the main caveat, but it has held up under daily use.

Gunnar Eriksson

Feb 23, 2026

Years in this space

I've evaluated a lot of these over the years. What stands out here is embedding and vector storage — handled better than most — and reduces boilerplate for developers. Quality depends on site structure is my one real gripe. Worth the time if this is your use case.

Nadia Petrova

Oct 16, 2025

Use it every day

Honestly didn't expect to like it this much. Embedding and vector storage is exactly what I needed, and reduces boilerplate for developers. I do wish quality depends on site structure, but I reach for it almost every day now and it just clicks.

Kwame Mensah

Oct 7, 2025

Skeptical, then convinced

I went in skeptical — most tools in this space overpromise. It actually delivers on automated web content extraction, and useful for docs and knowledge bases caught me off guard. Limited to website-based sources is why this isn't a perfect score, still, I'd recommend giving it a real trial.

问答

How does context stuff work?

Context stuff mode will scrape the current page, then store those results in your browser's storage. Then, each time you send a request using context stuffing mode, it will include the entire contents of the page in the system prompt.

Asked by Dovid Klein · May 11, 2026

How does retrieval work in followup questions?

Site RAG preforms retrieval (if you aren't using context stuffing mode) on each request by you. Once documents are retrieved, they are stored in your browser's storage with a sessionID tying them to your current chat session. Then, in followup questions the context from the most recent question will be included in the system prompt, and previous context will be passed as user messages before each of your previous requests, so the model always has the full context.

Asked by Uma Krishnan · May 1, 2026

What is "multi query mode"?

Multi query mode will generate multiple queries based on your input message, then preform a semantic search on each generated query. Finally, using all of the results, it will remove duplicates and generate a final response.

Asked by Julia Steiner · Feb 27, 2026

What happens if I index the same site twice?

Site RAG will ask you if you want to delete existing documents before indexing again, so you don't have to worry about duplicates. It deletes documents by URL.

Asked by Kenji Watanabe · Feb 18, 2026