Agent-Based Database Optimization SaaS (Public Preview)
DBScout is a vendor-neutral SaaS platform that uses a team of specialized agents to continuously scan and help database operations teams identify, log, understand and resolve potential PostgreSQL, MySQL, and MariaDB issues before they impact production systems.
Current supported databases
Connect Your Database and Get Insights Now
DBScout allows any operations team to benefit from our decade+ of PostgreSQL, MySQL, and MariaDB operational expertise by connecting to and pulling signals from your databases and leveraging AI to deliver DBA-grade insights and direction immediately.
Connect in Minutes: Attach read-only collectors to gather metadata and performance signals.
Receive Prioritized Findings: DBScout delivers severity-ranked issues, eliminating alert fatigue.
Human-in-the-Loop Approvals: DBScout is read-only by default, no direct changes to production.
How DBScout Works
Our agents continuously scan your database
Agents review the metadata, configuration settings, and performance signals DBScout collects from your databases for anomalies, up to every hour.
Specialists perform further analysis as needed
As scans reveal potential issues, specialist agents will perform deeper dives, e.g., the SysAdmin may analyze your compute if they notice redlining.
DBScout will log suggested interventions for approval
Agents will suggest actionable, evidence-based interventions that it will log into an approval queue for your review and change management.
DBScout’s Core Agents
The Senior DBA Lead
DB Scout’s lead operator. The Lead runs scheduled scans, triages findings, and decides what warrants deeper analysis. When needed, it routes cases to specialist agents, then synthesizes their outputs into a single, approval-ready remediation Pack. It focuses on clarity and trust: each Pack includes evidence, risk level, expected impact, safe steps, and rollback notes so teams can approve changes confidently and maintain an auditable record of decisions.
The Performance Specialist
Specializes in performance diagnosis and optimization. The Performance Engineer correlates workload symptoms with system constraints and database behavior to identify bottlenecks and likely root causes, e.g. contention, slow queries, I/O saturation, cache pressure, inefficient configurations. It produces evidence-backed recommendations with clear tradeoffs, expected gains, and rollback guidance, helping teams improve latency and throughput without guesswork.
The Security & Compliance Guardian
Focuses on security posture and auditability. The Security Auditor checks for common hardening gaps, risky configurations, and platform/engine concerns (including version/EOL and vulnerability-relevant signals where available). It prioritizes issues by severity and exposure, then produces remediation proposals detailing risk, evidence, validation steps, and rollback. This makes it easier to operationalize security work instead of relying on manual checklists.
The Capacity Architect
Helps teams stay ahead of resource growth and saturation. The Capacity Planner analyzes trends and headroom, e.g. storage growth, CPU/memory pressure, I/O patterns, to identify when capacity constraints are likely to become user-visible. It recommends scaling and tuning actions with practical thresholds, timelines, and risk considerations. The goal is to reduce capacity-driven incidents, giving lean teams clear, evidence-backed plans before system failure.
DBScout is Not Just O11y, It’s More
Dashboards only tell you that something is wrong. DBScout tells you exactly what to do. Every investigation results in a Remediation Brief — a comprehensive proposal designed for technical humans who need to make high-stakes decisions.
Evidence
Clear pointers to the logs, metrics, and configurations that triggered the finding.
Risk Rating
An honest assessment of the potential impact on your production environment.
Rollback Steps
Practical steps to revert changes if the environment doesn’t respond as expected.
Red Flag Detector
Candid advice on when to stay the course instead of intervening.
Key DBScout Concepts
Choose the Plan That Fits Your Needs
Starter
Ideal for single teams and early adopters
- 5 million tokens per collector
- Scans up to every 6 hours
- 7 day retention
- Team management
- RBAC
$29 / month / collector
Growth
Built for small platform teams with multiple databases.
- 10 million tokens per collector
- Scans up to every 2 hours
- 14 day retention
- Team management
- RBAC
$49 / month / collector
Business
The choice for larger teams with demanding needs.
- 20 million tokens per collector
- Scans up to every 1 hour
- 90 day retention
- Team management
- RBAC
$149 / month / collector
DBScout Agent Plan Availability
| AGENT | Starter | Growth | Business |
| Chat Agent |  | | |
| DBA Lead Agent | | | |
| Sysadmin Agent | | | |
| Performance Agent | | | |
| Query Plan Agent | | | |
| Security Agent | — | | |
| Backup Agent | — | | |
| Workload Profiler Agent | — | | |
| Capacity Planning Agent | — | — | |
Will my sensitive database records or PII (Personally Identifiable Information) be sent to your AI models?
No. DBScout is built on a privacy-first architecture. Our collectors are designed to gather metadata, configuration settings, and performance signals — not the actual data stored inside your tables. Furthermore, all signals are masked locally by the collector before being transmitted to our SaaS environment. This ensures that while our agents understand your database’s behavior and structure, they never see your actual customer data or sensitive records.
How does the “Analysis” credit system work compared to the “Regular Scans”?
Think of Scans as your automated, 24/7 background surveillance; they run on a set schedule (e.g., every 6 hours) to find common issues and surface “Findings.” An Analysis is a deeper, LLM-intensive “deep dive” triggered by a user or the DBA Lead Agent to solve a specific problem — like diagnosing a performance spike or analyzing a complex query plan. Each plan comes with a monthly token allowance.
Does DBScout require me to install ClusterControl or use Severalnines as my database hosting provider?
No. DBScout is a standalone, vendor-neutral SaaS product. It is designed to work with your existing database environment, whether it’s self-managed or you are a ClusterControl users. While it offers a natural upgrade path to our automation platforms, no other Severalnines software is required to use the platform.
Can DBScout make unauthorized changes to my production databases?
DBScout operates on a “read-only” basis by default, lacking the credentials or the mechanism to execute changes. You can enable auto-remediation, which creates a tightly scoped database user that can modify things like indexes and session settings. Otherwise, all remediation steps are delivered as reports for you to review, approve, and export to your own ticketing or orchestration systems.
What does Public Preview mean for DBScout?
Public Preview means the core platform capabilities and features are live and ready for testing in real-world environments. Although we’ll continuously be sanding down some rough edges and making enhancement, we’re providing access now to give our users practical value from AI while they help us shape our product roadmap. Feedback received during this phase is prioritized by our engineering team.