
WeKnora
Open-source knowledge framework for RAG Q&A, agent tasks and auto-built wikis
- Pricing
- Open source
- Platforms
- Web, self-hosted, API
- GitHub stars
- 37.6K
- Last commit
- 1 day ago
- Language
- Go
- Listing updated
- 3 days ago
About WeKnora
WeKnora is an open-source AI tool for knowledge and memory. It is available on Web, self-hosted, API and macOS.
WeKnora is a knowledge management framework from Tencent that combines retrieval-augmented question answering, agent reasoning and automatic wiki generation over a shared knowledge base. Answers link back to source passages so users can verify them.
Agents can use MCP tools and skills, run work in sandboxes (Docker, E2B, Cube) or a local browser, and cooperate in multi-agent setups. It syncs data from sources such as Feishu, Confluence and GitLab, and works with many model providers including OpenAI, Claude, Gemini, DeepSeek, Qwen and Ollama.
It is built for private deployment on Docker, Kubernetes or Helm, with a single-machine Lite edition and a desktop app, plus roles, API key scopes, OIDC and audit logs for teams.
What it does
- Hybrid retrieval Q&A
- Combines semantic and keyword search with multimodal embeddings and clickable source citations.
- Agent task execution
- Agents reason over multiple steps, call MCP tools and skills, and run in sandboxes.
- Automatic wiki
- Turns documents into linked wiki pages and a knowledge graph with version history.
- Multi-agent collaboration
- Splits work across several agents with separate conversations.
- Data source sync
- Imports files and syncs Feishu, Confluence, GitLab and similar sources.
- Team administration
- Offers space roles, API key scopes, OIDC sign-in and audit logs.
Good for
- Answer team questions from internal documents
- Maintain a wiki generated from raw documents
- Share a knowledge base with Claude or Cursor over MCP
- Automate scheduled research tasks with approvals
Pricing
Open source
Open source, self-hosted; license not stated in the provided content
Prices change. Check weknora.weixin.qq.com before you buy.
WeKnora FAQ
Can it be self-hosted?
Yes, it supports Docker, Kubernetes and Helm, plus a Lite single-machine edition and a desktop app.
Does it work with MCP?
Agents can call MCP tools, and the knowledge base can be exposed through an MCP server.
Which models can it use?
The site lists OpenAI, Claude, Gemini, DeepSeek, Qwen, Ollama, LiteLLM and several others.
Is WeKnora free?
Yes. WeKnora is open source, so you can use and self-host it for free.
What are the best alternatives to WeKnora?
The closest alternatives to WeKnora are ECC, ponytail, graphify, JavaGuide and claude-mem.
Alternatives to WeKnora

Skills, hooks, memory and security layer for coding agent harnesses
- Freemium
- Open-source agent
- 305K GitHub stars
- AI models and LLM tooling

Ruleset that makes AI coding agents write less code that still works
- Open source
- Open-source agent
- 192K GitHub stars
- AI models and LLM tooling

Code knowledge graph that gives coding assistants shared, source-linked context
- Freemium
- Open-source agent
- 149K GitHub stars
- Knowledge and memory

Chinese-language Java backend interview guide with an AI application section
- Free
- Open-source agent
- 160K GitHub stars
- Knowledge and memory

Memory for coding agents that saves decisions and fixes, with optional hosted sync
- Freemium
- Open-source agent
- 118K GitHub stars
- Knowledge and memory

Open-source RAG engine with a document pipeline, hybrid search and visual agent workflows
- Freemium
- Open-source agent
- 94.3K GitHub stars
- Knowledge and memory