Skip to content
Back to search

jonex

All-in-One Multimodal Parsing Engine + Ontology-Powered, LLM Wiki-Driven AI-Ready Knowledge Engine

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/yuezhiai/jonex with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/yuezhiai/jonex'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/yuezhiai--jonex/badge.svg)](/listings/yuezhiai--jonex)

HTML

<a href="/listings/yuezhiai--jonex"><img src="/api/listings/yuezhiai--jonex/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Docker Compose is the fastest way to run the complete platform.

Docker requirements

- Docker Engine or Docker Desktop - Docker Compose v2 with Buildx - `make` on macOS or Linux - Sufficient disk space and time for the first build, which downloads container images, Python dependencies, and RAG models

1. Clone the repository

git clone https://github.com/yuezhiai/jonex.git
cd jonex

2. Initialize configuration

make init

This creates:

- `deploy/.env` for the platform, database, object storage, and LLM Gateway - `deploy/.env.rag` for LightRAG, embeddings, and parsing - `deploy/.env.mcp` for the MCP server - Frontend `.env` files for Shell, Core Business, Ecosystem Management, Platform Management, and Dev Gateway

3. Configure model connections

Configure at least one OpenAI-compatible LLM and embedding provider in `deploy/.env`:

LLMGW_UPSTREAM_LLM_HOST=https://your-openai-compatible-host/v1
LLMGW_UPSTREAM_LLM_API_KEY=your_llm_api_key

LLMGW_UPSTREAM_EMBED_HOST=https://your-embedding-host/v1
LLMGW_UPSTREAM_EMBED_API_KEY=your_embedding_api_key

If your model names differ from the defaults, update the matching `LLM_MODEL` / `EMBEDDING_MODEL` keys — they are defined in both `deploy/.env` and `deploy/.env.rag`, and `EMBEDDING_MODEL` must be kept identical in the two files (it is used to build the vector index).

Keep `LIGHTRAG_API_KEY` identical in `deploy/.env` and `deploy/.env.rag`. For audio, video, or advanced image processing, also configure the VLM and ASR connections in `deploy/.env`.

4. Build and start

make build
make up
make ps

The first build creates the shared `jonex/python-base:local` image before Compose builds the platform services in parallel.

Use `make logs` to follow service logs when troubleshooting; press `Ctrl+C` to stop following them without stopping the platform.

5. Open Jonex

Visit:

http://localhost/

Local demo credentials:

Username: admin
Password: admin123
Tenant: tenant_jonex_demo

> **Security warning:** These credentials are for local evaluation only. Change or remove the default administrator account before binding Jonex to a non-loopback interface, sharing the deployment, or exposing it to any network. Complete the production checklist in [SECURITY.md](SECURITY.md) before deployment.

Windows PowerShell

.\jonex.ps1 init
.\jonex.ps1 build
.\jonex.ps1 up
.\jonex.ps1 ps

Use `.\jonex.ps1 logs` when you need to follow service logs.

If script execution is restricted:

powershell -ExecutionPolicy Bypass -File .\jonex.ps1 help

Stop the platform

make down

On Windows:

.\jonex.ps1 down

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 310 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. <p align="center">
  2. <img src="./docs/assets/logo-en.svg" alt="Jonex logo" width="320" />
  3. </p>
  4. <p align="center">
  5. <strong>AI Multimodal Domain Knowledge Engine</strong><br />
  6. Built for Enterprise Agents · Powered by the Dual Engines of Jonex Wiki and Graph Ontology
  7. </p>
  8. <hr />
  9. <br />
  10. <p align="center">
  11. <img src="https://img.shields.io/badge/Python-3.12-3776AB?logo=python&logoColor=white" alt="Python 3.12" />
  12. <img src="https://img.shields.io/badge/React-TypeScript-3178C6?logo=react&logoColor=white" alt="React and TypeScript" />
  13. <img src="https://img.shields.io/badge/FastAPI-009688?logo=fastapi&logoColor=white" alt="FastAPI" />
  14. <img src="https://img.shields.io/badge/Docker-Compose-2496ED?logo=docker&logoColor=white" alt="Docker Compose" />
  15. <img src="https://img.shields.io/badge/Multi--Tenant-Ready-0051D8" alt="Multi-tenant ready" />
  16. </p>
  17. <p align="center">
  18. English | <a href="README.zh.md">中文</a>
  19. </p>
  20. <p align="center">
  21. <a href="#overview">Overview</a> |
  22. <a href="#quick-start">Quick Start</a> |
  23. <a href="#local-development">Local Development</a> |
  24. <a href="#api-quickstart">API Quickstart</a> |
  25. <a href="#integration-profile">Integration profile</a> |
  26. <a href="#document-and-video-parsing">Document and video parsing</a> |
  27. <a href="#runtime-requirements">Runtime requirements</a> |
  28. <a href="#community-and-security">Community &amp; Security</a> |
  29. <a href="#license">License</a>
  30. </p>
  31. <p align="center">
  32. <strong>
  33. If you find value in Jonex, please ⭐ Star the project to support us.
  34. </strong>
  35. </p>
  36. ---
  37. ## Overview
  38. Jonex unifies an all-in-one multimodal parsing engine with an AI-ready knowledge engine. Ontology compiles domain reasoning into the knowledge layer before retrieval begins.
  39. It is an end-to-end enterprise AI knowledge platform that turns raw content into reusable knowledge services. Jonex connects data ingestion, multimodal parsing, domain knowledge compilation, vector and graph indexing, source-grounded retrieval, feedback loops, and business applications in one governed system.
  40. <p align="center">
  41. <img src="./docs/assets/jonex-knowledge-pipeline.png" alt="Jonex knowledge pipeline: from multimodal raw data to actionable knowledge" width="100%" />
  42. </p>
  43. ## Jonex Wiki
  44. Ontology+Wiki, one of Jonex's core engines, is powered by [Jonex Wiki](https://github.com/yuezhiai/jonex-wiki). Jonex Wiki is an LLM Wiki processing engine that follows the OKF format. As a sub-project spun out of Jonex, it offers semantic understanding and intelligent retrieval, knowledge structuring and automatic linking, natural language Q&A, continuous knowledge accumulation and collaborative updates, and full source traceability — lowering the barrier to knowledge access and improving knowledge reuse efficiency.
  45. Welcome to try it out! [Get started](https://github.com/yuezhiai/jonex-wiki)
  46. ## Quick Start
  47. Docker Compose is the fastest way to run the complete platform.
  48. ### Docker requirements
  49. - Docker Engine or Docker Desktop
  50. - Docker Compose v2 with Buildx
  51. - `make` on macOS or Linux
  52. - Sufficient disk space and time for the first build, which downloads container images, Python dependencies, and RAG models
  53. ### 1. Clone the repository
  54. ```bash
  55. git clone https://github.com/yuezhiai/jonex.git
  56. cd jonex
  57. ```
  58. ### 2. Initialize configuration
  59. ```bash
  60. make init
  61. ```
  62. This creates:
  63. - `deploy/.env` for the platform, database, object storage, and LLM Gateway
  64. - `deploy/.env.rag` for LightRAG, embeddings, and parsing
  65. - `deploy/.env.mcp` for the MCP server
  66. - Frontend `.env` files for Shell, Core Business, Ecosystem Management, Platform Management, and Dev Gateway
  67. ### 3. Configure model connections
  68. Configure at least one OpenAI-compatible LLM and embedding provider in `deploy/.env`:
  69. ```env
  70. LLMGW_UPSTREAM_LLM_HOST=https://your-openai-compatible-host/v1
  71. LLMGW_UPSTREAM_LLM_API_KEY=your_llm_api_key
  72. LLMGW_UPSTREAM_EMBED_HOST=https://your-embedding-host/v1
  73. LLMGW_UPSTREAM_EMBED_API_KEY=your_embedding_api_key
  74. ```
  75. If your model names differ from the defaults, update the matching `LLM_MODEL` / `EMBEDDING_MODEL` keys — they are defined in both `deploy/.env` and `deploy/.env.rag`, and `EMBEDDING_MODEL` must be kept identical in the two files (it is used to build the vector index).
  76. Keep `LIGHTRAG_API_KEY` identical in `deploy/.env` and `deploy/.env.rag`. For audio, video, or advanced image processing, also configure the VLM and ASR connections in `deploy/.env`.
  77. ### 4. Build and start
  78. ```bash
  79. make build
  80. make up
  81. make ps
  82. ```
  83. The first build creates the shared `jonex/python-base:local` image before Compose builds the platform services in parallel.
  84. Use `make logs` to follow service logs when troubleshooting; press `Ctrl+C` to stop following them without stopping the platform.
  85. ### 5. Open Jonex
  86. Visit:
  87. ```text
  88. http://localhost/
  89. ```
  90. Local demo credentials:
  91. ```text
  92. Username: admin
  93. Password: admin123
  94. Tenant: tenant_jonex_demo
  95. ```
  96. > **Security warning:** These credentials are for local evaluation only. Change or remove the default administrator account before binding Jonex to a non-loopback interface, sharing the deployment, or exposing it to any network. Complete the production checklist in [SECURITY.md](SECURITY.md) before deployment.
  97. ### Windows PowerShell
  98. ```powershell
  99. .\jonex.ps1 init
  100. .\jonex.ps1 build
  101. .\jonex.ps1 up
  102. .\jonex.ps1 ps
  103. ```
  104. Use `.\jonex.ps1 logs` when you need to follow service logs.
  105. If script execution is restricted:
  106. ```powershell
  107. powershell -ExecutionPolicy Bypass -File .\jonex.ps1 help
  108. ```
  109. ### Stop the platform
  110. ```bash
  111. make down
  112. ```
  113. On Windows:
  114. ```powershell
  115. .\jonex.ps1 down
  116. ```
  117. ## Local Development
  118. Local development uses root-level environment files and VSCode Debug. It is separate from the Docker deployment configuration under `deploy/`.
  119. ### Toolchain requirements
  120. - Python `>=3.12.13`
  121. - Node.js `>=20.18.0` (Node.js 22 LTS recommended)
  122. - pnpm `>=9.0.0`
  123. - A current stable version of uv
  124. ### Initialize the local environment
  125. ```bash
  126. cp .env.local.example .env.local
  127. cp .env.rag.local.example .env.rag.local
  128. mkdir -p .vscode
  129. cp docs/examples/launch.json.example .vscode/launch.json
  130. make frontends-install
  131. ```
  132. Set local middleware addresses in `.env.local` to `127.0.0.1`, or replace `SERVER_IP` with a remote infrastructure host. Backend processes are started from VSCode Run and Debug; the Makefile no longer starts host backend processes.
  133. Start the required local dependencies:
  134. ```bash
  135. make dev-infra-up # PostgreSQL, Redis, etcd, MinIO, and Milvus
  136. # Or, when running the complete RAG stack locally:
  137. make dev-deps-up # Middleware plus LightRAG and Atomic RAG
  138. ```
  139. Start the frontend gateway and applications in separate terminals:
  140. ```bash
  141. make dev-gateway
  142. make dev-frontend
  143. ```
  144. Open `http://localhost:8080`.
  145. ## Your First Knowledge Search in Five Minutes
  146. 1. Sign in with the local demo credentials `admin / admin123` in the demo tenant `tenant_jonex_demo`.
  147. 2. Open Core Business and create or select a domain space.
  148. 3. Create a knowledge base and organize it with folders or tags.
  149. 4. Select a parser profile or preset for the content you plan to ingest.
  150. 5. Upload files, or configure a REST API or S3-compatible data source.
  151. 6. Wait for multimodal parsing and knowledge compilation to finish.
  152. 7. Inspect the parsing results, compiled ontology, relationships, and knowledge graph.
  153. 8. Open Knowledge Search, ask a question, verify its references, and submit feedback.
  154. ## API Quickstart
  155. All external APIs are exposed through the unified Gateway.
  156. ### Sign in
  157. ```bash
  158. curl -X POST "http://localhost/api/v1/auth/login" \
  159. -H "Content-Type: application/json" \
  160. -H "X-Tenant-ID: tenant_jonex_demo" \
  161. -d '{"username":"admin","password":"admin123"}'
  162. ```
  163. Use the returned `access_token` to call ontology-first search:
  164. ```bash
  165. curl -X POST "http://localhost/api/v1/knowledge-base/search/ontology" \
  166. -H "Authorization: Bearer <access_token>" \
  167. -H "Content-Type: application/json" \
  168. -d '{
  169. "query": "What are the key risks described in these documents?",
  170. "knowledge_base_ids": ["<knowledge-base-id>"],
  171. "mode": "hybrid",
  172. "top_k": 5,
  173. "with_reasoning": true
  174. }'
  175. ```
  176. The response includes the answer, matched knowledge bases, ontology instances, RAG usage, structured source references, and an optional reasoning trace.
  177. In production, the browser communicates only with Frontend Gateway. Business APIs, capability services, and infrastructure components are not directly exposed to frontend applications.
  178. ## Integration profile
  179. - [RAG-Anything](https://github.com/HKUDS/RAG-Anything) and [MinerU](https://github.com/opendatalab/MinerU) can be connected for multimodal content processing and document parsing
  180. - [LightRAG](https://github.com/HKUDS/LightRAG) can be connected through the graph-enhanced retrieval adapter
  181. - [Neo4j](https://neo4j.com/) and [Milvus](https://milvus.io/) are packaged graph and vector persistence integrations
  182. - OpenAI-compatible endpoints provide replaceable LLM, embedding, reranking, VLM, and ASR services
  183. ## Document and video parsing
  184. - Deploy the Atomic RAG parser with Docker Compose, run it as an independently scaled capability, or register a compatible parser service through a parser profile
  185. - Process video locally with ASR, keyframes, and vision-language models, or route media analysis to a configured cloud service
  186. - Scale parsing workers independently from the core platform; GPU acceleration is optional for model-heavy workloads
  187. ## Runtime requirements
  188. | Deployment profile | Requirements |
  189. |---|---|
  190. | Core platform | Docker Engine or Docker Desktop, Docker Compose v2 with Buildx, PostgreSQL 15, Redis 7, and object storage |
  191. | Vector retrieval | Milvus, etcd, and MinIO or compatible equivalents |
  192. | Ontology graph | A supported graph database service; Neo4j is the packaged integration |
  193. | CPU parsing | Suitable for evaluation and light workloads; capacity scales with file size and concurrency |
  194. | Accelerated parsing | Optional NVIDIA GPU and Container Toolkit for faster OCR, ASR, and vision-language processing; VRAM depends on the selected models |
  195. | Cloud parsing | A compatible parsing or media-analysis endpoint, credentials, object storage, and outbound network access |
  196. ## Community and Security
  197. - Read [CONTRIBUTING.md](CONTRIBUTING.md) before opening an issue or pull request.
  198. - Participation is governed by [CODE_OF_CONDUCT.md](CODE_OF_CONDUCT.md).
  199. - Report vulnerabilities privately according to [SECURITY.md](SECURITY.md); do not include vulnerability details in a public issue.
  200. - See [CHANGELOG.md](CHANGELOG.md) for release notes and compatibility changes.
  201. ## License
  202. This repository is licensed under the [Jonex Open Source License](LICENSE), based on Apache License 2.0 with additional conditions. Third-party components remain under their respective licenses; see [NOTICE](NOTICE) and [THIRD_PARTY_NOTICES.md](THIRD_PARTY_NOTICES.md).
  203. ---
  204. <div align="center">
  205. ### **Jonex** — AI Multimodal Domain Knowledge Engine
  206. *Providing enterprise Agents with ontology-powered knowledge services that are reasoning-ready, traceable, and callable.*
  207. <br/>
  208. ### Enjoying Jonex?
  209. <a href="https://github.com/yuezhiai/jonex/stargazers">
  210. <img src="https://img.shields.io/github/stars/yuezhiai/jonex?style=social" alt="Stars"/>
  211. </a>
  212. <a href="https://github.com/yuezhiai/jonex/network/members">
  213. <img src="https://img.shields.io/github/forks/yuezhiai/jonex?style=social" alt="Forks"/>
  214. </a>
  215. **Click ⭐ Star to support us, thank you!**
  216. </div>
  217. © 2026 JONEX

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.