### OpenViking: The Context Database for AI Agents
English / [中文](README_CN.md) / [日本語](README_JA.md)
Website ·
Live Demo ·
GitHub ·
Issues ·
Docs
[](https://github.com/volcengine/OpenViking/releases)
[](https://github.com/volcengine/OpenViking)
[](https://github.com/volcengine/OpenViking/issues)
[](https://github.com/volcengine/OpenViking/graphs/contributors)
[](https://github.com/volcengine/OpenViking/blob/main/LICENSE)
[](https://github.com/volcengine/OpenViking/commits/main)
👋 Join our Community
📱
Lark Group ·
WeChat ·
Discord ·
X
***
## What is OpenViking
OpenViking is an open-source context database for AI agents. It stores memories, resources, and skills as one virtual filesystem under the `viking://` protocol, so an agent browses its own context with `ls`, `tree`, and `find` instead of querying a black-box vector store. Content is processed into three tiers — L0 abstract, L1 overview, L2 details — and loaded on demand. Every retrieval leaves a trajectory you can watch and debug. Full introduction: [Getting started](https://docs.openviking.ai/en/getting-started/01-introduction).
[](https://openviking.ai/studio)
*The [OpenViking Studio](https://openviking.ai/studio) playground — a live demo you can open in the browser, no installation required.*
## Why OpenViking
- **One filesystem for all context.** Memories, resources, and skills each get a `viking://` URI. Agents locate and manipulate context deterministically, like a developer working with files. → [Viking URI](https://docs.openviking.ai/en/concepts/04-viking-uri) · [Context types](https://docs.openviking.ai/en/concepts/02-context-types)
- **Tiered loading cuts token spend.** Every entry is processed into L0 (abstract), L1 (overview), and L2 (details) on write, then loaded only as deep as the task requires. → [Context layers](https://docs.openviking.ai/en/concepts/03-context-layers)
- **Directory recursive retrieval.** Vector search first locates the highest-scoring directory, then drills down layer by layer, so results arrive with their surrounding context intact. → [Retrieval](https://docs.openviking.ai/en/concepts/07-retrieval)
- **Observable retrieval.** Each query preserves its directory-browsing trajectory. When a result looks wrong, you can see exactly which path produced it. → [Retrieval](https://docs.openviking.ai/en/concepts/07-retrieval)
- **Sessions become memory.** After a session commits, OpenViking asynchronously extracts user preferences and agent experience into long-term memory. → [Session](https://docs.openviking.ai/en/concepts/08-session)
How the pieces fit together: [Architecture](https://docs.openviking.ai/en/concepts/01-architecture). The thinking behind the design: [The Database Paradigm for Context Engineering](https://blog.openviking.ai/post/openviking-context-database/).
```
viking://
├── resources/ # Resources: project docs, repos, web pages, etc.
│ └── my_project/
│ ├── docs/
│ │ ├── api/
│ │ └── tutorials/
│ └── src/
└── user/
└── {user_id}/
├── memories/
│ └── preferences/
│ ├── writing_style
│ └── coding_habits
├── resources/
│ └── private_project/
├── skills/
│ ├── search_code
│ └── analyze_data
└── peers/
└── web-visitor-alice/
```
The three loading tiers:
- **L0 (Abstract)**: a one-sentence summary for quick relevance checks.
- **L1 (Overview)**: core information and usage scenarios for planning.
- **L2 (Details)**: the full original data, read only when needed.
Each directory carries its own L0/L1 layers, so relevance can be judged before any full file is read:
```
viking://resources/my_project/
├── .abstract # L0: ~100 tokens - quick relevance check
├── .overview # L1: ~2k tokens - structure and key points
└── docs/
├── .abstract
├── .overview
└── api/
├── auth.md # L2: full content, loaded on demand
└── endpoints.md
```
## Proof it works
OpenViking 0.3.22 has been evaluated on long-conversation user memory (LoCoMo) and multi-turn agent tasks (tau2-bench). Full results and setup details, including knowledge-base QA, are in the [benchmark report](https://blog.openviking.ai/post/openviking-benchmark-results/); reproduction scripts live in [./benchmark](./benchmark).
