tencent/weknora is an open source LLM knowledge platform written in Go. It bundles into a single repository RAG (retrieval-augmented generation) that lets you query source documents, a reasoning agent that handles multiple steps on its own, and a Wiki that organizes and maintains knowledge. The latest release is v0.8.3, and the repository was created on July 22, 2025.
The center of this repository is understanding and indexing documents, and providing semantic search plus document navigation with attached evidence. Document understanding is the stage that reads the source and turns it into a form that can be searched. The README explains that you can gather team documents in one place, search them and keep them up to date, get answers that verify evidence through RAG on the same knowledge base, run multi-step tasks with the Agent, and organize knowledge with the Wiki. The project's design is that the same set of documents is reused for search, task execution, and knowledge organization. The repository has 34,840 GitHub stars.
That said, no changelog is given for what changed in v0.8.3. There are no figures showing retrieval accuracy or processing performance either. The default model, the embedding approach, and the configuration needed for deployment are not included. The license is judged as NOASSERTION, so the license type cannot be confirmed. The only evaluation score filled in is the momentum score of 8.5, and the novelty, technology, and quality items have no values. Bundling three features into one repository is broad in scope, but the license file and the retrieval quality on an actual document set are the parts that need separate checking. With the record at hand, all that can be said is the scope of the structure.