English | 简体中文 | UpdateLog | Help Docs
ChatWiki is a workflow automation platform focused on the WeChat ecosystem, dedicated to making every official account a super AI agent. It fully integrates the open capabilities of the official account platform, allowing you to build WeChat ecosystem applications through drag-and-drop, enabling features like one-click rewriting of official account articles and AI-curated comment replies.
-
Industry First: Automatic reply to private messages for unverified official accounts, supporting text, voice, images, mini-program cards, video messages, etc.
-
WeChat Workflows: Integrates trigger scenarios such as user private messages, comments, follows, unfollows, menu clicks, etc. Supports various processing steps like replying to private messages, tagging fans, generating draft articles, publishing articles, and more.
-
Knowledge Base Synchronization: Supports scraping articles and materials from official accounts to build a knowledge base with one click.
-
Workflow Orchestration: Conversational workflows, plugin workflows, including basic workflow nodes, bidirectional MCP, Agent mode, and user interaction.
-
Document Knowledge Base: Supports URL reading, batch document import, API integration, AI-based segmentation, QA segmentation, parent-child segmentation. Supports knowledge graphs, hybrid vector search, and visual exploration of knowledge graphs.
-
QA Knowledge Base: Automatically extracts QA knowledge from uploaded documents, supports automatic clustering of unknown questions, and summarizes common FAQs from human conversations.
-
Human Handoff: Handles general user inquiries via bot, while also supporting human staff. Issues that the bot cannot resolve can be escalated to human客服, with multi-agent collaborative assignment.
-
Model Support: Supports over 20 mainstream global models, including DeepSeek R1, doubao pro, qwen max, OpenAI, Claude, etc.
-
Multiple Deployment Options: Offers desktop client, supports publishing as a WebApp, and embedding into websites, official accounts/service accounts, wecome staff, WeChat store staff, etc.
-
MCP & API Integration: Allows integration of external MCP services or publishing workflows as MCP services. Complete OpenAPI interface for easy integration with existing business systems.
-
Multi-account Permission Management: Three-tier permission system (admin, editor, viewer) ensures data isolation. IP whitelist and permanent login logs.
- 🌍 Free Trial URL: chatwiki.com
- 🖼️ Screenshots:
ChatWiki Community Edition is deployed using Docker and can be installed in just a few simple steps:
# Install Docker
sudo curl -sSL https://get.docker.com/ | CHANNEL=stable sh
# Clone the project
git clone https://github.com/zhimaAi/chatwiki.git
cd chatwiki/docker
# Start the service
docker compose up -d
# Start using it, access via IP:port (ensure the specified port ${CHAT_SERVICE_PORT}, default 18080, is open)
# Default username: admin
# Default password: chatwiki.com@123
For any issues or suggestions during installation and deployment, please contact us or refer to the help documentation for assistance. You can also check the guides below.
-
Frontend: vue.js
-
Backend: golang + python
-
Database: PostgreSQL16 + pgvector + zhparser
Feel free to contact us for help or to provide suggestions to improve ChatWiki. You can reach us through:
- Help: Check the help documentation
- Email: Send an email to jarvis@2bai.com.cn to contact us.
- WeChat: Scan the QR code below to join the ChatWiki tech community group. Please add the note "chatwiki" when adding.
For the complete changelog, please click 👉️👉️UpdateLog.md
2026/08/14
1.[STD] Cloud version custom models support time-based pricing (compatible with DeepSeek's new pricing policy)
2.BookToSkill: Optimized functionality, added update logic, and added online OCR support (via Alibaba Cloud API)
3.Bot prompts: Added support for inserting the current time system variable
4.[STD] Added the Agent official website (https://chatwikiagent.xiaokefu.com.cn)
5.Thinking is disabled by default for all models with optional thinking unless explicitly enabled by the user
2026/08/07
1.External Service: Bot supports LINE app
2.Agent: My Skills interface interaction adjustments and logic optimization
3.Temporarily disabled DeepSeek official API thinking mode
4.Agent: Local document reading skill now supports docx, csv, and pdf files
5.Fixed issue where customer messages were not displayed in unknown question session context
6.Fixed issue where Alibaba Cloud OCR document parsing results were not using pagination query logic
7.Alibaba Cloud OCR document parsing: Enabled FormulaEnhancement logic
8.Fixed time field error in Alibaba Cloud OCR document parsing task status query
9.[STD] Added Alibaba Cloud OCR document parsing open API and credit validation
10.Agent skill generation tool: Web to Skill now supports skill updates
11.Knowledge base: Added validity period settings (permanent, specified date, specified days)
2026/07/31
1.External Service: Embed website SDK supports initialization parameters
2.Optimized data table integer/number field data storage and rendering
3.Fixed issue where variable list dropdown was not working when inserting variables in bot prompts
4.External Service: Added popup size and position settings for embedded website
5.Agent local knowledge base query tool: Support LLM autonomous selection of which knowledge base to query
6.Agent bot, chat bot, and workflow: Added rate limiting settings
7.[STD] Associated ChatWiki bot: Additionally forward image, voice, and video messages when multimodal is enabled
8.Agent bot: Optimized associated knowledge base
9.Fixed issue where chat test and external service user messages were obscured when rate limiting had no reply
10.[STD] Cloud version supports DingTalk, WeCom, and Feishu login
This project follows the ChatWiki Open Source License. The ChatWiki Open Source License is based on the Apache License 2.0, but with additional restrictions:
- ChatWiki is free for individual users, including non-commercial or commercial activities conducted by individuals.
- Any company, organization, institution, or team that uses ChatWiki for commercial purposes must contact us to obtain a commercial license.
- When using ChatWiki's frontend components, you may not remove or modify the "ChatWiki" logo, trademark, or copyright notice contained therein.
The full license text can be found in the LICENSE file. For commercial licensing, please contact us.