OpenViking
OpenViking is an open-source context database designed specifically for AI agents.
- Type
- Agent Skill
- Open source
- Yes
- GitHub Stars
- ★ 26.0k
- Source
- skill-github
- Repository
- github.com/volcengine/OpenViking
Overview
OpenViking is an open-source context database designed for AI agents (such as openclaw). It unifies the management of agent context (memory, resources, and skills) through a filesystem paradigm, enabling hierarchical context passing and self-evolution. Developers can build an agent's 'brain' just like managing local files, solving context fragmentation issues, reducing token consumption, improving retrieval effectiveness, and supporting visualized retrieval traces and automatic session management.
Capabilities
- ▪Unified management of memory, resources, and skills
- ▪Hierarchical context loading to reduce token consumption
- ▪Directory-based recursive retrieval to improve retrieval effectiveness
- ▪Visualized retrieval traces to observe context
- ▪Automatic session management for context self-iteration
Use cases
Setup
pip install openviking --upgrade --force-reinstall
This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.
FAQ
Which operating systems does OpenViking support?
Supports Linux, macOS, and Windows.
What model capabilities does OpenViking require?
Requires VLM models and Embedding models.
Related skills
mantis
Provides security review capabilities for AI coding agents, automatically discovering, reproducing, and fixing vulnerabilities.
sast-skills
Transform your AI coding assistant into a SAST scanner to automatically detect vulnerabilities in code.
Lightswind-UI-Library
AI-native CLI-first React component library with MCP Server support.
scientific-agent-skills
Transform any AI agent into a scientific assistant with 147 ready-to-use research skills.
susi_alexa_skill
A skill that enables question-and-answer interactions between Alexa and Susi AI.
claude-context
Provides code search capabilities for Claude Code, turning the entire codebase into context.