chaitin/PandaWiki

PandaWiki 是一款 AI 大模型驱动的开源知识库搭建系统，帮助你快速构建智能化的产品文档、技术文档、FAQ、博客系统，借助大模型的力量为你提供 AI 创作、AI 问答、AI 搜索等能力。

📚 What is PandaWiki?

PandaWiki is an open‑source, AI‑powered wiki platform. It lets you create product docs, technical manuals, FAQs, or blogs and then augments them with large‑language‑model capabilities such as:

AI‑assisted authoring – the model can draft or improve content.
AI‑driven Q&A – visitors can ask natural‑language questions and get answers drawn from the wiki.
AI‑enhanced search – semantic retrieval powered by the same model.

All of this runs on a self‑hosted Docker stack, but a SaaS version is also offered.

⚡ Core Features

Feature	What it does
AI‑assisted creation	Write in Markdown/HTML; the model can suggest text, fill sections, or rewrite content.
AI Q&A & search	Users type questions; the backend queries the indexed docs with a large model to return concise answers.
Rich export	Docs can be exported as Word, PDF, or plain Markdown.
Third‑party integration	Embed the wiki as a widget, or expose it through DingTalk, Feishu, WeChat Work bots, etc.
Content import	Pull in pages from URLs, sitemaps, RSS feeds, or upload offline files.
Multi‑knowledge‑base	Each knowledge base gets its own public wiki site, isolated from others.

🛠️ Getting Started

Prerequisite – a Linux server with Docker 20.x+ and root access.
Install – run the one‑liner:
```
bash -c "$(curl -fsSLk https://release.baizhi.cloud/panda-wiki/manager.sh)"
```
The script walks you through configuration and starts the containers.
Login – after installation you’ll see the internal/external URLs, admin username, and password.
Configure a model – the first login prompts you to connect a large‑model service. You can use the provided “Baizhi Cloud Model Marketplace” (free 5 CNY credit on sign‑up) or enter your own API keys manually.
Create a knowledge base – via the admin console, define a new KB, upload or import documents, and let the system “learn” them.
Use the wiki – the generated front‑end site lets visitors read docs, search, or ask questions.

🎯 Typical Use Cases

Product documentation – keep manuals up‑to‑date and let customers ask natural‑language questions.
Internal tech knowledge base – employees can retrieve answers without digging through pages.
FAQ / support portal – AI answers reduce support load.
Blog platform with AI assistance – writers get drafting help and readers can query the content.

🤝 Community & Contributions

Pull requests and issues are welcomed on GitHub.
A WeChat group is provided for real‑time help.
Documentation (installation, model integration, troubleshooting) is hosted at pandawiki.docs.baizhi.cloud.

📄 License

PandaWiki is released under GNU Affero General Public License v3.0 (AGPL‑3.0) – you may use, modify, and redistribute the code, but any network‑served modifications must also be open‑sourced under the same terms.