openviking
OpenViking context database for AI agents — filesystem-paradigm context management with tiered loading (L0/L1/L2), hierarchical retrieval, and automatic session memory. Alternative/complement to flat RAG for agent context.
Best use case
openviking is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
OpenViking context database for AI agents — filesystem-paradigm context management with tiered loading (L0/L1/L2), hierarchical retrieval, and automatic session memory. Alternative/complement to flat RAG for agent context.
Teams using openviking should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/openviking/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How openviking Compares
| Feature / Agent | openviking | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
OpenViking context database for AI agents — filesystem-paradigm context management with tiered loading (L0/L1/L2), hierarchical retrieval, and automatic session memory. Alternative/complement to flat RAG for agent context.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# OpenViking — Context Database for AI Agents > Open-source context management system using filesystem paradigm for AI agents. > Source: https://github.com/volcengine/OpenViking | Stars: 6K+ ## Core Concept OpenViking treats agent context like a filesystem (`viking://` protocol): - **L0**: One-sentence summary (always loaded) - **L1**: ~2K token overview (loaded on demand) - **L2**: Full details (loaded when deep dive needed) Reduces token consumption vs loading full context every turn. ## Installation ```bash pip install openviking --upgrade --force-reinstall ``` Requirements: Python 3.10+, Go 1.22+, C++ compiler **Config** (`~/.openviking/ov.conf`): ```toml [embedding] provider = "openai" # or volcengine, litellm [vlm] provider = "openai" model = "gpt-4o" ``` ## Key CLI Commands ```bash # Add a resource to the context DB ov add-resource ./docs/architecture.md --tags "architecture,backend" # Search context ov find "authentication patterns" ov grep "database schema" # Start VikingBot agent ov bot start ``` ## Architecture ``` viking:// ├── memories/ # Session memories (auto-extracted) ├── resources/ # Documents, code, specs ├── skills/ # Agent capabilities └── preferences/ # User/project preferences ``` ## Tiered Retrieval Flow 1. Lock high-score directory (vector search over L0 summaries) 2. Refine: load L1 overviews of top directories 3. Zoom in: load L2 full content only when needed 4. Track: every retrieval trajectory is logged (transparent RAG) ## vs UltraThink Memory | Feature | UltraThink | OpenViking | |---------|-----------|-----------| | Storage | Neon Postgres + pgvector | Filesystem + vector index | | Retrieval | Category/scope SQL | Hierarchical directory | | Context tiers | Single level | L0/L1/L2 progressive | | Session mgmt | Custom hooks | Built-in auto-extract | | Language | TypeScript | Python + Go | ## Integration Pattern OpenViking can serve as the **resource/document** layer for UltraThink — store large reference docs in OpenViking for efficient tiered retrieval while UltraThink handles episodic memory (solutions, patterns, decisions). ```bash # Index UltraThink skills into OpenViking ov add-resource .claude/skills/ --recursive --tags "skills" ov add-resource memory/src/ --tags "memory,typescript" ```
Related Skills
ultrathink
UltraThink Workflow OS — 4-layer skill mesh with persistent memory and privacy hooks for complex engineering tasks. Routes prompts through intent detection to activate the right domain skills automatically.
ultrathink_review
Multi-pass code review powered by UltraThink's quality gate — checks correctness, security (OWASP), performance, readability, and project conventions in a single structured pass.
ultrathink_memory
Persistent memory system for UltraThink — search, save, and recall project context, decisions, and patterns across sessions using Postgres-backed fuzzy search with synonym expansion.
ui-design
Comprehensive UI design system: 230+ font pairings, 48 themes, 65 design systems, 23 design languages, 30 UX laws, 14 color systems, Swiss grid, Gestalt principles, Pencil.dev workflow. Inherits ui-ux-pro-max (99 UX rules) + impeccable-frontend-design (anti-AI-slop). Triggers on any design, UI, layout, typography, color, theme, or styling task.
Zod
> TypeScript-first schema validation with static type inference.
webinar-registration-page
Build a webinar or live event registration page as a self-contained HTML file with countdown timer, speaker bio, agenda, and registration form. Triggers on: "build a webinar registration page", "create a webinar sign-up page", "event registration landing page", "live training registration page", "workshop sign-up page", "create a webinar page", "build an event page", "free webinar landing page", "live demo registration page", "online event page", "create a registration page for my webinar", "build a training event page".
webhooks
Webhook design patterns — delivery, retry with exponential backoff, HMAC signature verification, payload validation, idempotency keys
web-workers
Offload heavy computation from the main thread using Web Workers, SharedWorkers, and Comlink — structured messaging, transferable objects, and off-main-thread architecture patterns
web-vitals
Core Web Vitals monitoring (LCP, FID, CLS, INP, TTFB), measurement with web-vitals library, reporting to analytics, and optimization strategies for Next.js
web-components
Native Web Components, custom elements API, Shadow DOM, HTML templates, slots, lifecycle callbacks, and framework-agnostic design patterns
wasm
WebAssembly integration — Rust to WASM with wasm-pack/wasm-bindgen, WASI, browser usage, server-side WASM, and performance considerations
vue
Vue 3 Composition API, Nuxt patterns, reactivity system, component architecture, and production development practices