Skip to content

V.connect Fremdaufruf — Overview

V.connect Fremdaufruf lets a KIS open a pre-filled V.ap product form (DUBA, ELIM+, …) in the workstation's browser, even when the KIS itself cannot make the HTTPS POST call V.ap normally expects. It runs adjacent to the KIS on the workstation, accepts the GET URL the KIS knows how to emit, and bridges to V.ap's POST-based session initialisation.

This document is the customer-direction overview: what the component is for, what it gives you, and what adopting it involves. For the deeper architecture and design rationale, see the Architecture. For the KIS-integrator workflow of generating URL templates, see the V.ap Fremdaufruf URL-Builder.


1. Purpose

V.ap product modules expose POST endpoints that accept structured JSON (patient identification, report data, court-message references, …) and respond with a one-time login link to a pre-filled form. KIS systems that can make HTTPS POST calls integrate with these endpoints directly.

Many KIS configurations cannot. The integration mechanism they expose is a single primitive: "open this URL, with these parameters appended". The naïve workaround — accept the same payload via GET on a V.ap endpoint — fails on data protection: URLs leak into browser history, into proxy logs, into server access logs, and into the Referer header of any link the user clicks afterwards.

Fremdaufruf closes this gap on the workstation side. The component runs locally, accepts the GET URL the KIS emits, transforms it into the authenticated POST V.ap expects, and 303-redirects the browser to the resulting login link. The V.ap modules themselves are unchanged; only the local component is added.


2. At a glance

Where Fremdaufruf sits — on the workstation, between the KIS and V.ap Hospital network · KIS workstation KIS application + embedded browser V.connect Fremdaufruf single static binary Vertama Cloud V.ap GET URL HTTPS POST
On the workstation, Fremdaufruf receives the KIS's GET URL and translates it into the authenticated HTTPS POST V.ap expects. The browser then follows V.ap's redirect to the pre-filled form.

What the customer gets:

  • A single small binary on each KIS workstation, or one container on the hospital LAN — running as a Windows service or Linux container. Single-digit MB memory, no runtime dependencies.
  • Compatibility with KIS systems that have only the basic "open external URL" capability — no KIS-side development needed.
  • End-user experience preserved: the pre-filled form renders inside the KIS's embedded browser, no window switching, no separate login.
  • Module-agnostic: one component, all V.ap modules. Adding a module to the integration is a configuration change, not a new deployment.

3. Integration independence

The integration mechanism Fremdaufruf relies on — opening an external URL with parameters appended — is one of the most universal KIS capabilities. Almost every clinical information system supports it; it predates any vendor-specific integration API. By relying on this primitive, integrations with V.ap modules don't require the KIS vendor to provide module-specific support, build custom adapters, or unlock paid integration tiers. The hospital chooses what to integrate; the KIS vendor neither gates nor prices that decision.


4. Deployment shapes

V1 supports two shapes from the same binary, distinguished by configuration:

Per-workstation, loopback-bound. Each KIS workstation runs its own instance bound to 127.0.0.1. Operationally simple — one workstation, one install. Trust boundary: the workstation operating system. Distributed as a signed Windows installer that registers a native Windows service.

Hospital-network central. A single installation on the hospital LAN, bound to a routable address, serves multiple KIS workstations. Trust boundary: a configured access secret that every caller must supply on every request. Distributed as a container image at ghcr.io/mcp-health/v.connect/fremdaufruf (linux/amd64), with a Docker Compose example as a starting point.

Both shapes use the same binary, the same configuration schema, and the same upstream contract with V.ap. The choice is a deployment decision, not a product decision.


5. What adopting looks like

5.1 Prerequisites

Item Requirement
Platform Windows 10 or later (x64) for per-workstation deployments; Linux x86-64 via container (Docker / Podman) for central deployments. macOS is development-only.
Outbound network HTTPS reachability from the workstation (or LAN host) to the V.ap base URL.
V.ap API user Provisioned by your Vertama contact, with the module scopes the integration requires.
Service-installation rights Administrator (Windows) or root (Linux) for service registration. Day-to-day operation runs unprivileged.
Disk Less than 50 MB for the binary, configuration, and audit log.

The component is a single static binary with no runtime dependencies — no JRE, no .NET runtime, no shared libraries beyond the OS itself.

5.2 Installation paths

  • Windows. Signed installer registers the Windows service, drops the binary in %PROGRAMFILES%, and writes a configuration template. Authenticode signature is verifiable before install.
  • Linux / container. Container image with a Docker Compose example as a starting point. Configuration via environment variables (12-factor) or a mounted config file.
  • macOS. Development only.

5.3 Operationally

Once installed, the component runs as a long-running service and exposes a browser-based admin dashboard at http://127.0.0.1:8811/admin on the host. The administrator sets the admin access code, edits the configuration to point at the V.ap instance, reloads the configuration from the dashboard, and confirms the V.ap connectivity test passes. Installation completes in minutes; first successful KIS call typically same day once configuration is in place.

Day-to-day operation is hands-off. The dashboard shows live activity counters, the effective configuration (with secrets redacted), and the recent request history.

The complete operations manual — installation procedure, full configuration schema, troubleshooting, audit-log details — ships with the binary as the Betriebshandbuch and is reachable from the admin dashboard. Vertama supplies it on request prior to deployment for IT-security review.


6. Security & compliance — summary

Operator sovereignty. Once installed, the Fremdaufruf component runs entirely within your control. Vertama has no remote admin surface, no phone-home, no inbound channel into your deployment. All configuration is local. The component's network footprint is exactly what you configure: outbound only to the V.ap base URL you set (it does not reach into your intranet even if your network would permit it), and inbound only on the bind address and port you choose.

What Vertama provides and where it runs:

Artifact Where it runs Controlled by
Fremdaufruf component (Windows native service) KIS workstation Hospital admin
Fremdaufruf component (Linux container) Hospital-managed host Hospital admin
V.ap product modules + URL-builder Vertama cloud Vertama operates; hospital admin provisions and consumes via API user

Three independent controls govern what can be done through Fremdaufruf:

  • V.ap-side authorisation (primary) — the configured API user's scope at V.ap governs which V.ap actions are reachable at all.
  • Local path allowlist (defense in depth) — hospital admins may narrow the surface to a list of permitted upstream paths.
  • Access secret (intranet trust boundary) — required for non-loopback deployments; every request must supply the matching _s parameter.

Wire-level: HTTPS to the configured V.ap base URL only; TLS verification against the system trust store; parameter values and keys are never logged at any log level (only the count); credentials and access secrets are never logged.

For the full security posture — authority model, wire-level guarantees per deployment shape, trust boundaries, audit-log guarantees — see Architecture §8.


7. Support

Vertama supports Fremdaufruf directly. The team behind it is small, which means: when you reach us, you reach engineering, not a ticket queue. Response is fast for installation-blocking issues; functional improvements roll into the regular release cadence.


8. What's in V1 — and what's not

In V1:

  • Both deployment shapes (per-workstation loopback, hospital-network intranet).
  • Signed Windows installer and Linux container distribution.
  • Browser-based admin dashboard for status, configuration reload, and V.ap-connectivity testing.
  • TOML configuration with environment-variable overrides.
  • Stderr and file audit-log destinations.
  • URL-builder tooling at V.ap for the customer's chosen modules (DUBA, ELIM+, and other V.ap modules use the same URL-builder workflow at no additional component cost).

For what is deliberately not in scope today, with rationale, see Architecture §9.


Companion documents:

  • Architecture — full design rationale, deployment-shape detail, security posture, scope and roadmap.
  • V.ap Fremdaufruf URL-Builder — KIS-integrator guide to generating URL templates.
  • Betriebshandbuch — operations manual; ships with the binary, reachable from the admin dashboard after install.

Last updated: 2026-06-11 Applies to: V.connect Fremdaufruf V1