After eighteen months of design-partner work, Insightral is generally available. Starting today, any team can deploy the advisor to their private cloud, connect their database estate, and receive ranked findings within minutes.
What ships today
- 99 deterministic advisory rules across PostgreSQL, Oracle, MySQL, and SQL Server — no AI in the critical path.
- SHA-256 fingerprint-only egress: query literals never leave your agent host.
- Bring-your-own-AI: route optional explanations through OpenAI, Anthropic, Azure, AWS Bedrock, or Google Vertex inside your environment. Your keys, your negotiated rates, your audit log.
- 60+ MCP tools so Claude, Cursor, or any agentic client can query findings and runbooks from your editor.
- Availability monitors, status pages, and a correlation service map from endpoint to finding to fix.
Deployment
Insightral runs as a Docker Compose stack or a Kubernetes deployment in the customer private cloud. The collector talks to your databases over read-only credentials. The control plane — findings, baselines, alerts, and the AI router — stays in your network. Nothing routes through Valora infrastructure.
Private-cloud and BYO-AI
We built Insightral for teams whose security posture makes a shared SaaS database advisor a non-starter. Private cloud is not an add-on tier — it is the only deployment model. The AI layer is optional and uses provider keys you already hold: no Valora-hosted inference ever touches customer database context.
Pricing
Licensing is scoped to your estate. There is no self-serve checkout. Email sales@valorasolutionsinc.com with a description of your environment and we respond within one business day. A 30-day Proof of Value is available by request.
What is next
The roadmap covers deeper Oracle advisory rules, a Kubernetes-native collector, and expanded MCP coverage. Design partner relationships for Q4 are open — reach out if you want to influence direction.