OpenDerisk

OpenDeRisk

OpenDeRisk is an AI-Native Risk Intelligence System designed as your application system’s intelligent manager, providing 7×24 hour comprehensive and in-depth protection.

Features

1. DeepResearch RCA: Quickly locate root causes through in-depth analysis of logs, traces, and code.
2. Visualized Evidence Chain: Fully visualize diagnostic processes and evidence chains for clear, accurate judgment.
3. Multi-Agent Collaboration: SRE-Agent, Code-Agent, ReportAgent, Vis-Agent, and Data-Agent working in coordination.
4. Open-Source Architecture: Built with a completely open architecture, enabling framework and code reuse in open-source projects.

Architecture

Introduction

The system employs a multi-agent architecture. Currently, the code primarily implements the highlighted components. Alert awareness is based on Microsoft’s open-source OpenRCA dataset. The decompressed dataset is approximately 26GB. On this dataset, we achieve root cause analysis through multi-agent collaboration, with Code-Agent dynamically writing code for final analysis.

Technical Implementation

Data Layer: Pull the large-scale OpenRCA dataset (20GB) from GitHub, decompress locally, and process for analysis.

Logic Layer: Multi-agent architecture with SRE-Agent, Code-Agent, ReportAgent, Vis-Agent, and Data-Agent collaborating for deep DeepResearch RCA (Root Cause Analysis).

Visualization Layer: Use the Vis protocol to dynamically render the entire processing flow and evidence chain, as well as the multi-role collaboration and switching process.

Digital Employees (Agents) in OpenDeRisk

Install (recommended)

Install via curl

# Download and install latest version
curl -fsSL https://raw.githubusercontent.com/derisk-ai/OpenDerisk/main/install.sh | bash

Configuration File

After installation, the default configuration file is automatically initialized at: ~/.openderisk/configs/derisk-proxy-aliyun.toml

Edit this file and set your API keys:

vi ~/.openderisk/configs/derisk-proxy-aliyun.toml

Start

openderisk-server  

From source(development)

Install uv (required)

macOS/Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows:

powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

Clone and Install Dependencies

git clone https://github.com/derisk-ai/OpenDerisk.git

cd OpenDerisk

# Install Dependencies with uv
uv sync --all-packages --frozen \
    --extra "base" \
    --extra "proxy_openai" \
    --extra "rag" \
    --extra "storage_chromadb" \
    --extra "derisks" \
    --extra "storage_oss2" \
    --extra "client" \
    --extra "ext_base" \
    --extra "channel_dingtalk"

Note: channel_dingtalk is optional. Skip it if you don’t need DingTalk channel support.

Start Server

🚀 Quick Start (Zero Configuration, Recommended)

Start without any configuration file:

# Method 1: Use quickstart command
uv run derisk quickstart

# Method 2: Use startup script
./start.sh

# Method 3: Specify port
uv run derisk quickstart -p 8888

After starting, visit http://localhost:7777 and configure models and settings through the web UI.

For detailed instructions, see: Quick Start Guide

📝 Start with Configuration File

Configure the API_KEY in derisk-proxy-aliyun.toml, then run:

Note: By default, we use the Telecom dataset from OpenRCA. Download via: gdown https://drive.google.com/uc?id=1cyOKpqyAP4fy-QiJ6a_cKuwR7D46zyVe

After downloading, move datasets to pilot/datasets/

Run the startup command:

# Start with configuration file
uv run derisk quickstart -c configs/derisk-proxy-aliyun.toml

# Or use traditional method
uv run python packages/derisk-app/src/derisk_app/derisk_server.py --config configs/derisk-proxy-aliyun.toml

Access Web UI

Open your browser and visit http://localhost:7777

Usage Modes

• AI-SRE (OpenRCA)
  • Notice: We use the OpenRCA Dataset Bank Dataset
  • Download: gdown https://drive.google.com/uc?id=1enBrdPT3wLG94ITGbSOwUFg9fkLR-16R
  • Place datasets in ${derisk}/pilot/datasets
• Flame Graph Assistant
  • Upload flame graphs (Java/Python) from your local application for analysis
• DataExpert
  • Upload metrics, logs, traces, or Excel data for conversational analysis

Development

• Agent Development
  • Refer to implementations under derisk-ext.agent.agents
• Tool Development
  • Skills
  • MCP (Model Context Protocol)
• DeRisk-Skills
  • derisk-skills

Execution Results

Citation

If you find this repository helpful, please cite:

@misc{di2025openderiskindustrialframeworkaidriven,
      title={OpenDerisk: An Industrial Framework for AI-Driven SRE, with Design, Implementation, and Case Studies}, 
      author={Peng Di and Faqiang Chen and Xiao Bai and Hongjun Yang and Qingfeng Li and Ganglin Wei and Jian Mou and Feng Shi and Keting Chen and Peng Tang and Zhitao Shen and Zheng Li and Wenhui Shi and Junwei Guo and Hang Yu},
      year={2025},
      eprint={2510.13561},
      archivePrefix={arXiv},
      primaryClass={cs.SE},
      url={https://arxiv.org/abs/2510.13561}, 
}

Acknowledgement

• DB-GPT
• GPT-Vis
• MetaGPT
• OpenRCA

The OpenDeRisk-AI community is dedicated to building AI-native risk intelligence systems. 🛡️ We hope our community can provide you with better services, and we also hope that you can join us to create a better future together. 🤝

Community Group

Join our DingTalk group and share your experience with other developers!