Alibaba Cloud RDS Copilot Intelligent Operations Assistant

Scenario Description

Alibaba Cloud CLI + RdsAi OpenAPI

• Understand user's natural language requests (Chinese or English), identify if related to RDS Copilot
• Directly call Alibaba Cloud CLI to execute

  aliyun rdsai chat-messages

  command for real-time RDS Copilot queries
• When receiving results or user-pasted error messages, further explain, diagnose, and provide recommendations

Installation

Pre-check: Alibaba Cloud CLI must be installed

This skill uses Alibaba Cloud CLI to call RdsAi OpenAPI. You need to install and configure Alibaba Cloud CLI first.

macOS Installation

# Option 1: Install via Homebrew (recommended)
brew install aliyun-cli

# Option 2: Install via PKG package
curl -O https://aliyuncli.alicdn.com/aliyun-cli-latest.pkg
sudo installer -pkg aliyun-cli-latest.pkg -target /

# Option 3: Install via one-click script
/bin/bash -c "$(curl -fsSL https://aliyuncli.alicdn.com/install.sh)"

Linux Installation

# Install via one-click script
/bin/bash -c "$(curl -fsSL https://aliyuncli.alicdn.com/install.sh)"

# Or download TGZ package for manual installation
curl https://aliyuncli.alicdn.com/aliyun-cli-linux-latest-amd64.tgz -o aliyun-cli.tgz
tar xzvf aliyun-cli.tgz
sudo mv aliyun /usr/local/bin/

Verify Installation

aliyun version

Credential Configuration

Option 1: Interactive Configuration (Recommended)

aliyun configure --profile rdsai

• Access Key Id: Your AccessKey ID
• Access Key Secret: Your AccessKey Secret
• Default Region Id: cn-hangzhou (or other regions)

Option 2: Non-interactive Configuration

aliyun configure set \
  --profile rdsai \
  --mode AK \
  --access-key-id <yourAccessKeyID> \
  --access-key-secret <yourAccessKeySecret> \
  --region cn-hangzhou

Command Format

Basic Command Structure

aliyun rdsai chat-messages \
  --query '<query content>' \
  --inputs RegionId=<region ID> Language=<language> Timezone=<timezone> [CustomAgentId=<custom agent ID>] \
  --event-mode separate \
  --endpoint rdsai.aliyuncs.com \
  --user-agent 'AlibabaCloud-Agent-Skills' \
  [--conversation-id '<conversation ID>']

Parameter Description

IMPORTANT: Parameter Confirmation — Before executing any command, Determine user intent: SQL writing/optimization, SQL diagnosis, instance parameter tuning, troubleshooting, performance analysis, query instance list, etc. Collect necessary parameters (use default values if not specified).

--query

--inputs RegionId=

cn-hangzhou

--inputs Language=

zh-CN

--inputs Timezone=

Asia/Shanghai

--inputs CustomAgentId=

--event-mode

separate

--endpoint

rdsai.aliyuncs.com

--conversation-id

--region

--profile

--user-agent

AlibabaCloud-Agent-Skills

RAM Permissions

rdsai:ChatMessages

Core Workflow

1. Confirm Task Type and Parameters

• RegionId

  : Region ID (default

  cn-hangzhou

  )

• Language

  : Language (default

  zh-CN

  )

• Timezone

  : Timezone (default

  Asia/Shanghai

  )

• CustomAgentId

  : Custom Agent ID (optional)

• --conversation-id

  : Conversation ID for multi-turn dialogue (optional)

2. Construct Command and Call CLI

# Basic query
aliyun rdsai chat-messages \
  --query 'List RDS MySQL instances in Hangzhou region' \
  --inputs RegionId=cn-hangzhou Language=zh-CN Timezone=Asia/Shanghai \
  --event-mode separate \
  --endpoint rdsai.aliyuncs.com \
  --user-agent 'AlibabaCloud-Agent-Skills'

# Troubleshooting example
aliyun rdsai chat-messages \
  --query 'RDS instance rm-bp1xxx connection timeout, error Too many connections, please help troubleshoot. Instance is in Hangzhou region.' \
  --inputs RegionId=cn-hangzhou Language=zh-CN Timezone=Asia/Shanghai \
  --event-mode separate \
  --endpoint rdsai.aliyuncs.com \
  --user-agent 'AlibabaCloud-Agent-Skills'

# Query with Beijing region
aliyun rdsai chat-messages \
  --query 'Optimize this SQL: SELECT * FROM users WHERE name LIKE "%test%"' \
  --inputs RegionId=cn-beijing Language=zh-CN Timezone=Asia/Shanghai \
  --event-mode separate \
  --endpoint rdsai.aliyuncs.com \
  --user-agent 'AlibabaCloud-Agent-Skills'

# Multi-turn dialogue (using ConversationId from previous response)
aliyun rdsai chat-messages \
  --query 'Continue analyzing the above issue' \
  --conversation-id '<ConversationId from previous response>' \
  --inputs RegionId=cn-hangzhou Language=zh-CN Timezone=Asia/Shanghai \
  --event-mode separate \
  --endpoint rdsai.aliyuncs.com \
  --user-agent 'AlibabaCloud-Agent-Skills'

# Using custom Agent
aliyun rdsai chat-messages \
  --query 'Analyze database performance' \
  --inputs RegionId=cn-hangzhou Language=zh-CN Timezone=Asia/Shanghai CustomAgentId=your-custom-agent-id \
  --event-mode separate \
  --endpoint rdsai.aliyuncs.com \
  --user-agent 'AlibabaCloud-Agent-Skills'

3. Parse Results and Follow-up Processing

• Explain RDS Copilot's response to the user in natural language
• If the response contains SQL or operational steps, assess risks and warn:
  • Avoid executing high-risk statements directly in production (e.g., large table

    DELETE

    /

    UPDATE

    / schema changes)
  • Recommend validating in test environment or adding backup/condition restrictions
• If continuing the conversation, record the

  ConversationId

  from the response for the next query

Output Format

{"data":{"ConversationId":"8227be22-xxxx-xxxx-xxxx-xxxxxxxxxxxx","Event":"workflow_started","MessageId":"a79c881c-xxxx-xxxx-xxxx-xxxxxxxxxxxx",...}}
{"data":{"Answer":"<partial answer content>","Event":"message",...}}
{"data":{"Event":"workflow_finished",...}}

• ConversationId

  : Conversation ID (for multi-turn dialogue)

• Answer

  : AI assistant's response content

• Event

  : Event type (workflow_started, message, workflow_finished)

Success Verification

1. CLI installation successful:

   aliyun version

   shows version number
2. Credential configured correctly:

   aliyun configure list

   shows configured credentials
3. API call successful: Response contains

   ConversationId

   and

   Answer

   in JSON format
4. Response content valid: Answer is relevant to the query content

Cleanup

API and Command List

aliyun rdsai chat-messages

Best Practices

1. Use multi-turn dialogue: For complex issues, use

   --conversation-id

   for context-aware multi-turn conversations
2. Specify correct region: Set

   RegionId

   parameter based on the RDS instance's region
3. Be cautious in production: SQL recommendations from RDS Copilot should be validated in test environment first
4. Save conversation ID: Save the returned

   ConversationId

   if you need to follow up or continue analysis
5. Use configuration file: Recommend using

   aliyun configure

   to configure credentials, avoid exposing sensitive information in command line
6. Use --profile: You can configure multiple credential profiles and switch between accounts using

   --profile

Reference Links