openbkn-ai / bkn-foundry

# BKN Foundry
中文 | English
🌐 **Website:** openbkn.ai
OpenBKN is an ontology-driven business knowledge network platform. Through ontology modeling, it turns the data and logic scattered across documents, systems, processes, rules, and expert experience into a business knowledge network that agents can understand, execute, and verify — so agents land accurately, safely, and reliably in real business environments, not just generating answers but continuously creating executable, traceable, and verifiable business value.
**BKN Foundry** is the technical foundation of OpenBKN, providing that business knowledge network with unified data access, safe execution, and governance.
**On this page:** 📚 Quick links · 🚀 Quick start · 🛠️ OpenBKN SDK · 🛡️ Administration · 🏗️ BKN Foundry · 📐 BKN Lang · 📊 Benchmarks
> **Note:** BKN Foundry is a **backend-only framework** — it does not include a web UI. All interactions are through the CLI, SDK, or API.
## 📚 Quick Links
- 🛠️ OpenBKN SDK - End-user / agent BKN CLI, TypeScript SDK, and the AI agent skill
- 🤝 Contributing - Guidelines for contributing to the project
- 🚢 Deployment - One-click deploy to Kubernetes
- 📘 Documentation - Product documentation and usage guides (EN / 中文)
- 📖 API Reference - OpenAPI docs for every service (YAML source + rendered Markdown / interactive HTML)
- 📦 Samples - Runnable end-to-end walkthroughs, in their own repo
- 🧾 Release Notes - All notable changes
## 🚀 Quick Start
1. **Prerequisites & planning** — read the Deployment Guide and satisfy its prerequisites. **Linux** is the supported target for full installs; **macOS** local dev (kind) is optional — see Mac install (dev) (中文).
2. **Pre-install host check / fix with `preflight.sh`** (recommended)
 On the **target install host**, run a system check before `deploy.sh`. It verifies kernel / sysctl / containerd / `kubectl` / `helm` / Node / BKN CLIs and can fix what's missing (each fix is opt-in unless `-y`):
 [code] 
 Exit codes: **0** OK, **1** any FAIL, **2** only WARN. Use `--report=/tmp/preflight.txt` to keep a full log.
3. **Run installation scripts**:
 [code] 
4. **Verify the deployment**:
 [code] 
5. **Post-install bootstrap with `onboard.sh`** (recommended)
 On the **same host** as the install (where `kubectl` reaches the cluster), run the post-install bootstrap. It (re-runnable) registers an LLM + an embedding, patches the BKN ConfigMaps when the default embedding actually changes, and on a **full install** also creates the business user **`test`**, assigns every role from `openbkn admin role list`, and switches `openbkn` to that user:
 [code] 
 > **Why `sudo`?** `onboard.sh` reads `$HOME/.openbkn-ai/config.yaml` (written by `sudo deploy.sh` into `/root/.openbkn-ai/`) and writes the `openbkn` auth token to `$HOME/.bkn`. Running it without `sudo` falls back to the in-repo template `conf/config.yaml` and may resolve a different access URL. **macOS dev path** (`bash dev/mac.sh onboard`) does **not** need `sudo`.
 Re-runs are safe: