Automate Game Customer Support with Alibaba Cloud SLS Intelligent Q&A Assistant

Background

Game operation teams receive massive daily complaints across five main scenarios: payment, item loss, matchmaking, anti‑cheat appeals, and social disputes. Investigating each issue traditionally requires 3‑6 manual log queries across multiple SLS Logstores, leading to slow resolution, heavy reliance on expert knowledge, and backlog during peak periods.

Goal

Build an AI‑driven automated troubleshooting system based on an SOP knowledge base that can automatically query logs, determine root causes, and draft standard replies.

Solution: SLS Intelligent Q&A Assistant

The assistant enables natural‑language queries to locate answers in massive logs, shortening the customer‑service cycle.

Core Capabilities

Automatic Log Query : Recognizes the scenario and executes SOP‑defined SLS queries, reducing investigation time from minutes to seconds.

Automatic Root‑Cause Identification : Compares key fields and outputs precise conclusions for 13 predefined scenarios.

Automatic Reply Generation : Combines the root cause with SOP scripts to produce a draft response for agent approval.

Overall Architecture

The system consists of a digital employee (AI agent), an SOP knowledge base, and the SLS log service. The agent queries logs based on the SOP, determines the root cause, and generates a response.

Step‑by‑Step Implementation

1. Create a Digital Employee

Log in to the Alibaba Cloud Log Service console (

https://account.aliyun.com/login/login.htm?oauth_callback=https://sls.console.aliyun.com/lognext/profile

), locate the SOP Q&A Assistant, and create a new digital employee. Fill in basic information and later attach the SOP knowledge base.

2. Prepare the SOP Knowledge Base

Write SOP documents that define the supported business scenarios. An example SOP file is available at

https://github.com/aliyun/sop-chat/blob/main/sop-docs-example/game-sop/sop-docs/SOP.md

. The SOP determines the scope of the AI assistant.

3. Host the SOP in Cloud Code (Codeup)

Create a Git repository on Cloud Code ( https://codeup.aliyun.com/) and push the SOP directory:

cd game-sop-docs</code>
<code>git init</code>
<code>git remote add origin [email protected]:619e222864fa260377314dee/game-sop-docs.git</code>
<code>git add .</code>
<code>git commit -m "init: add game sop docs"</code>
<code>git push -u origin HEAD

Record the Enterprise ID and Repository ID from the repository URL, and generate a personal access token from the user settings.

4. Link the SOP to the Digital Employee

In the digital employee’s knowledge‑base settings, add the Cloud Code repository configuration (Enterprise ID, Repository ID, Access Token) so the assistant can fetch the SOP files.

5. Test the Assistant with Sample Scenarios

Thirteen test cases covering the five complaint categories are provided. Example:

1. 你好，我今天在微信充了 98 元礼包，微信那边已经扣款成功了，但游戏里钻石一直没到账，等了很久了，请帮我查一下！账号：`U_BDD640FB`，区服：`cn_s18`，充值金额：98 元，微信支付，渠道订单号：`WX5912226647794000`

Enter the query in the assistant; it automatically runs the corresponding SLS query, identifies the root cause, and suggests a response.

6. Integrate with Messaging Platforms (DingTalk example)

Create a DingTalk robot via the DingTalk Open Platform, obtain the Client ID and Client Secret, and configure the robot in the assistant’s channel settings.

Deploy the sop-chat server (download from https://github.com/aliyun/sop-chat), grant execution permission, and start it:

# chmod +x sop-chat-server</code>
<code>./sop-chat-server

The server provides a UI for configuring Alibaba Cloud AccessKey credentials and channel information.

After saving, the robot can be searched in DingTalk (e.g., GameSupportDemo ) and used to answer player queries automatically.

Conclusion

By ingesting SOP knowledge into the SLS Intelligent Q&A Assistant and leveraging collected game logs, operators can answer business‑level questions quickly, greatly improving support efficiency and service quality. The assistant’s effectiveness depends on the completeness of the SOP documentation; richer SOPs expand the range of supported scenarios.