Memory, knowledge retrieval and skills usually live in three systems that never agree. OpenViking publishes itself as the layer that unifies them, and then keeps evolving what it holds.
An agent that queries three systems has to decide which one is authoritative when they disagree. Merging them moves that decision out of the prompt.
Filing capability next to memory and documents is the unusual half of this claim. Most stacks treat skills as code, not as something the database owns.
No latency figures, no recall scores, no corpus sizes. The trending description publishes none, so this page prints none.
Concept mock, not affiliated with the project. The three source boxes and the unified plate are drawn from the repository's own one-line description on GitHub Trending, 18 August 2026: "Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills." The diagram is marked illustrative because no schema or architecture drawing is published on that page.