Skip to content

[request]: Skill for database auditing automation (RDS SQL Server & Aurora PostgreSQL) #106

Description

@bbhavini0502

Select the type of tool this is related to

  • Skill
  • Custom agent
  • MCP server
  • IaC (Infrastructure as Code) template
  • Other

Tell us about your request

A skill that guides DevOps Agent through deploying, configuring, operating, and troubleshooting an AI-powered database auditing automation solution for Amazon RDS SQL Server and Aurora PostgreSQL — covering audit log collection, anomaly detection, and compliance report generation.

Which AWS service(s) is this for?

mazon RDS, Amazon Aurora, Amazon Bedrock, AWS Lambda, Amazon S3, Amazon EventBridge, Amazon SNS

What problem are you trying to solve?

Teams running RDS SQL Server and Aurora PostgreSQL need continuous audit logging, anomaly detection, and compliance reporting (SOX, PCI-DSS, HIPAA), but wiring up pgAudit / SQL Server native audit, streaming logs to storage, and turning them into meaningful, actionable findings is manual and error-prone. When an operator asks DevOps Agent to set up database auditing, generate a compliance report, or investigate suspicious privileged-user activity, the agent has no packaged, opinionated procedure to follow, so results are inconsistent and incomplete. Without this skill, operators fall back to piecing together the pipeline by hand and interpreting raw audit logs themselves.

How should it behave?

Triggers when a user asks about setting up database auditing on RDS SQL Server or Aurora PostgreSQL, deploying the audit automation solution, generating compliance reports from audit logs, detecting anomalies in database activity, monitoring privileged user access, or troubleshooting the audit log pipeline. It walks through: deploying the solution via CloudFormation, configuring pgAudit (Aurora PostgreSQL) or SQL Server native audit, enabling CloudWatch Logs export, and operating/troubleshooting the anomaly-detection and report-generation flow (Lambda + Bedrock → S3/SNS).

Which DevOps Agent functionalities is this relevant to?

Chat tasks

Does this need extra permissions?

No response

Are you currently working around this?

oday this is done manually — configuring audit extensions/options by hand, standing up the log pipeline per environment, and interpreting raw audit logs without a consistent, agent-driven procedure.

Have you checked for existing tools?

  • I've searched existing issues, skills, custom agents, and MCP servers, and this isn't already covered or extendable.
  • I've confirmed DevOps Agent can't already do this natively (I tested it).

Would you like to build it?

  • I'm willing to contribute this myself (see CONTRIBUTING.md).
  • I'm suggesting it for someone else to build.

Additional context

Source repository: https://github.com/aws-samples/sample-database-auditing-automation

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    needs-triageNew issue awaiting maintainer triagerequestCommunity request for a new tool (roadmap intake)skillRelates to a DevOps Agent skill

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions