
WeKnora
β Text source-checked Nov 22, 2026Tencent's open-source enterprise knowledge management framework for aggregating team materials for Q&A, task execution, and Wiki organization.
Source-checked against the vendor's site Β· rebuilt from the vendor's own pagesProfile updated Link checked How we rate
WeKnora is an open-source enterprise knowledge management framework from Tencent that aggregates team materials for knowledge Q&A, task execution, and Wiki organization. The vendor says that the platform uses RAG for querying materials, Agents for multi-step tasks, and Wiki for organizing knowledge, with all three capabilities sharing the same knowledge base.
The RAG mode combines semantic and keyword retrieval to find text and uses multimodal embeddings to find images, allowing users to click citations in answers to locate the original text for verification. The Agent mode reads attachments on demand, retrieves knowledge, calls MCP tools and skills, and executes tasks in a sandbox or local browser while supporting appending requirements, forking, or rolling back during conversations. The Wiki mode generates interlinked Wiki pages and knowledge graphs from raw documents, supporting browsing, editing, and version rollback.
WeKnora supports deployment via Docker, Kubernetes, and Helm, and offers a Lite single-machine version and a desktop application, with models, vector databases, and storage backends being replaceable as needed. The system synchronizes materials from platforms such as Feishu, Confluence, and GitLab, and allows querying and usage through IM, browser plugins, MCP Server, or API.
Access permissions are managed through space roles, resource sharing, API key scopes, and OIDC identity integration, with the ability to restrict services to access only whitelisted domains.
According to WeKnora: WeKnora is an open-source enterprise knowledge management framework from Tencent that aggregates team materials for knowledge Q&A, Task execution, and Wiki organization. Its pages list this strength: Supports flexible deployment via Docker, Kubernetes, Helm, Lite single-machine, and desktop apps with replaceable models and storage backends.
β The AI Tool Bible editorial team
Pros
- β Supports flexible deployment via Docker, Kubernetes, Helm, Lite single-machine, and desktop apps with replaceable models and storage backends.
- β Provides synchronization for Feishu, Confluence, and GitLab, with access via IM, browser plugins, MCP Server, or API.
- β Manages access permissions through space roles, resource sharing, API key scopes, and OIDC integration.
- β Allows users to click citations in answers to locate the corresponding original text for verification.
Facts
- Platforms
- api (not source-checked)
- API available
- Yes (not source-checked)
- Open source
- Yes
- Model used
- Multi-model (not source-checked)