WeKnora is an open-source Go-based knowledge platform that turns enterprise documents into searchable RAG, an autonomous reasoning agent, and a self-updating wiki.
WeKnora is an open-source, LLM-powered framework for enterprise document understanding, semantic retrieval, and reasoning. It is designed around three ways to use the same knowledge base: asking questions with RAG, running multi-step tasks through an agent, and organizing information in a wiki that stays aligned with the underlying documents. The README also positions it as a deployable knowledge platform for teams that want to keep data in their own environment.
It addresses the common enterprise problem of scattered documents that are hard to search, reason over, and keep current. Instead of treating search, automation, and documentation as separate systems, WeKnora aims to connect them to the same knowledge bases so teams can retrieve answers, delegate tasks to an agent, and maintain a living wiki from one place.
Conceptually, WeKnora ingests documents from supported sources, converts them into knowledge bases, and exposes that knowledge through retrieval, agent, and wiki experiences. The RAG path is for looking things up, the agent path is for multi-step reasoning and actions, and the wiki path is for structuring and curating knowledge; all three share the same underlying content. The README also says it supports editing, diffing, and rolling back retrieval chunks, cross-session memory, connected tools and sandboxes for agents, multiple model providers, and deployment choices ranging from local Docker to Kubernetes and cloud setups.
It is gaining attention because it sits at the intersection of several high-demand areas: RAG, agentic workflows, semantic search, vector search, enterprise knowledge management, and multi-tenant deployment. The repository also presents a broad feature set in one package—document connectors, many model vendors, IM integrations, tracing, permissions, and multiple deployment modes—which makes it appealing to teams looking for an end-to-end knowledge platform rather than a narrow demo. Its strong weekly star growth suggests active interest in that combined proposition.
The README does not name direct competitors, so only broad comparisons are clear. Conceptually, it overlaps with open-source RAG and knowledge-base systems, agent platforms, semantic search tools, and wiki/knowledge-management products; the distinguishing point in the README is that it combines those roles on top of the same knowledge base. It also integrates with external ecosystems such as OpenAI, Ollama, LiteLLM, and common collaboration or document platforms, which suggests it is meant to sit alongside rather than replace every source system.
AI-explained · grounded in each repo's README