Sentinel by NewHeap Pilot preview

Your existing applications.
Now AI-ready.

All applications running on a supported SQL Server or PostgreSQL database can become AI-ready for your organization with Sentinel. Your AI agents use their data through controlled access.

No new or additional application development.

We configure the gateway and access policies. For supported database versions, drivers and features. Pilot preview, not yet released for production.

ERP · Your existing application
Customers · Orders · Reports
1048North€12,4001049North€8,2001050North€4,200
SQL Server / PostgreSQL
SentinelRoles · permissions · security · logging
LI
Your user & AI agent

“What is the revenue in the North region?”

Your existing software, with Sentinel as the access layer.
SentinelPilot previewLI

Your existing application keeps working

A question. A check. The right response.

Choose a question and follow it through Sentinel.
01

Existing application

ERP · Sales administration
File · Edit · Reports
OrdersCustomersHR
Orders
OrderRegionRevenue
1048North€12,400
1049North€8,200
1050North€4,200
SQL Server / PostgreSQL

Your existing software and database. No additional application development.

02

Sentinel

Check before data access
User
Lisa
Role
Sales
Department
North
  • Read own region revenue
  • Protect customer fields
  • Block salary data
Roles · permissions · security

Sentinel checks the request and which data may be returned.

03

User & AI agent

LI
Lisa · Sales

What is the revenue in the North region?

AllowedREQ-01

€24,800

Revenue · North region

You may read revenue for your own region.

Only released information reaches the user and the selected AI model.

Every decision is traceable

Decision log
Decisions by request
RequestIdentityDecisionApplied policy
REQ-01user-07 · SalesAllowedRead permission
REQ-02user-07 · SalesProtectedField protection
REQ-03user-07 · SalesDeniedRole boundary

The audit log stores decisions and metadata, not question text, SQL, parameter values or result data.

  1. 1Ask a question
  2. 2Check permissions
  3. 3Response & log

Your existing software takes on a new role

AI-ready without
rebuilding your applications.

Your organization already uses applications for orders, customers or inventory. Sentinel makes data in their supported databases available to AI agents without new or additional application development. An agent can then prepare a report after a meeting using current business data. You decide which information the agent may use and which actions require approval.

Provide the access that is needed.

Define which tables and columns each application and user may access. Sentinel evaluates the complete query, including supported queries that combine several tables.

  • Access profiles for users and AI applications.
  • Rules for data within the organization.
  • Support within the declared SQL Server and PostgreSQL matrices.

Protect data before it is transferred.

The gateway checks results against configured policy. Configured source columns are also recognized through aliases and supported database views.

  • Release permitted values, mask, pseudonymize or block them.
  • When inspection is mandatory, a result that cannot be checked is refused.
  • Protection uses configured classifications and column rules; automatic detection of all personal data is not a current product feature.

Changes need a deliberate decision.

Reading data and changing it have different consequences. Where policy requires it, a supported write waits for separate human approval.

  • Separate approval for supported INSERT, UPDATE and DELETE statements.
  • Permissions are checked again after approval.
  • Approval expires and can only be used once.

Trace the decision. Keep the content private.

The audit log shows who made a request, which rules applied and what was decided. The protected content itself is excluded from this evidence log.

  • No SQL text, parameter values or result values in the audit log.
  • Search and export decisions and approvals.
  • Configure evidence retention and deletion.

Two ways to connect

For agents and
existing software.

For your AI agents

Through MCP or HTTP

Agents query databases through controlled access. Identity, permissions, result policies and any required approval are evaluated in the same chain.

For existing applications

Through the database connection

The application keeps its existing database and driver. For supported features, only the connection address changes to point to the gateway. Sentinel makes permitted data available to AI; this requires no application development.

Deployment with your IT team

Start within your
own environment.

In the customer-hosted preview, the connector and database gateway run within your network. Database credentials and raw results stay there; only results released according to policy reach the requester.

  • The connector initiates its connection to the control plane from inside the network.
  • Nexus is the gateway; Sentinel manages policy, approvals and evidence.
  • MCP and HTTP provide access without dependence on one AI model.
  • Where permitted results are subsequently processed depends on the chosen agent, model and deployment.

The NewHeap-hosted variant is still in development and is not available as a production service.

A focused pilot

Validate first.
Then build further.

Sentinel is a pilot preview by NewHeap. Supported features have been tested in development and pilot environments; there is no production release yet. Together, we define the database, access policies, test scenarios and acceptance criteria for your use case.

What do we review before a pilot?
  • Supported database versions, drivers, queries and result limits.
  • Masking and pseudonymization based on configured rules; automatic content classification is not yet available as a product feature.
  • Identity integrations, access revocation, recovery, performance and scale must be assessed for the chosen deployment.
  • Guidance, setup and costs are discussed separately. Sentinel is not included by default in the €28 subscription.
Your data stays yours

You retain ownership, access and control.

Data stored in Europe

Processing varies by service and selected model.

ISO 27001-compliant infrastructure

Security is part of the foundation.

Model independent

Choose suitable models, with Local AI where needed.