← Home

SECURITY AND TRUST

Security: what stays with you, what leaves

Beams of light running through a data centre, like data looking for a way outAgents in your environment · synthetic payloads only

The essentials in 30 seconds

  • Agents in your environment, orchestration as an Enforcis SaaS; on-premises deployment possible, subject to assessment, for critical environments. Hybrid topologies (your private cloud) are under study.
  • Test payloads are synthetic; no real customer business data is used as a campaign payload.
  • The test payloads that leave are synthetic decoys, sent to test destinations defined with you during scoping. Technical metadata (agent state, host and network information, campaign results) reaches the orchestration.
  • Your vendor assessment needs more than this page: we answer in writing.

What runs in your environment

  • Lightweight agents on Windows machines (signed binary; other systems: ask us), deployed in your own environment, one execution point per exit path you want to test.
  • Orchestration as an Enforcis SaaS; on-premises deployment possible, subject to assessment, for critical environments. Hybrid topologies (your private cloud) are under study.
  • In an on-premises deployment, logs and results stay in your environment. Upcoming versions will bring SIEM integrations and a sealed evidence layer.
  • Test payloads are synthetic; no real customer business data is used as a campaign payload.

What leaves, and where it goes

An exfiltration test only means something if something actually tries to leave. That is why the test payloads that leave are decoys, by design: synthetic data built to look real enough to trigger your rules (synthetic IBANs, synthetic HR files, “Confidential” labels), with no value of its own.

  • The test destinations are defined with you when scoping the POC or the pilot. The payloads that transit through them are synthetic.
  • Besides the test payloads, the agent’s control and collection flow carries technical metadata (agent state, host and network information, campaign results) to the orchestration; the exact dictionary, flows and ports are in the deployment documentation.
  • The scope (machines, channels, decoy categories) is defined with you before any campaign in production.
  • If you prefer to start small, the proof of concept runs one standard scenario with synthetic data from a representative machine in your network, through your real egress controls.

How decoys are designed is detailed in Synthetic data: testing exfiltration in production without using any real business data.

What version 1.0 covers

Version 1.0 tests 8 channels: HTTPS, HTTP, DNS, FTP, ICMP, GitHub Gist, Google Drive and DPaste. It does not replay removable media (USB), printing, OT networks, outbound email or native Microsoft 365 locations. The full list, and the controls usually involved for each channel, is on the Platform page.

Your vendor security assessment

Bringing a testing tool into your information system rightly calls for more than this page. As part of your vendor assessment, we answer in writing questions such as:

  • What privileges do the agents need, depending on the execution mode (service or standalone binary)?
  • How are the agents, the orchestrator and the console updated? Are the binaries signed?
  • How do users authenticate to the console, and which roles exist?
  • Which network flows does a campaign generate: ports, protocols, endpoints for each channel?
  • Which logs and results are produced, where are they kept, and what reaches the Enforcis cloud when orchestration runs there?
  • How are vulnerabilities in the platform handled?
  • Which subcontractors are involved, and under which data processing agreement?

These are also the questions we recommend asking any vendor in Choosing a DLP testing tool: criteria and questions for the vendor.

Vendor security review

Send us your questionnaire.

Choose “Vendor security review” as the subject of the demo form: we come back to you with written answers.

This website

  • Audience measurement with Matomo, without cookies, with truncated IP addresses and data hosted on our own server.
  • Forms (demo, kit) are processed by our processor HubSpot; the form itself states what is collected and why.
  • Forms are protected against automated submissions by an internal check, with no third-party service and no cookie.

Details are in the privacy notice. For a security question about Enforcis or this website, write to contact [at] enforcis.com.

Frequently asked questions

Is Enforcis a SaaS service?

Yes by default. Agents in your environment, orchestration as an Enforcis SaaS; on-premises deployment possible, subject to assessment, for critical environments. Hybrid topologies (your private cloud) are under study.

Does Enforcis see our data?

Test payloads are synthetic; no real customer business data is used as a campaign payload. No real business data is used as a payload. Technical metadata (agent state, host and network information, campaign results) reaches the orchestration, which runs as an Enforcis SaaS or, subject to assessment, on your premises. The exact data dictionary, hosting and retention are detailed in writing during your vendor assessment.

Where do the decoys go during a test?

To test destinations defined with you when scoping the POC or the pilot. Decoys leave by design: that is how a test shows what actually gets out, on the scenarios actually tested, so you can read it against your logs.

How do we get the answers our vendor assessment needs?

Ask for a vendor security review: choose that subject in the demo form and we answer your questions in writing.

What actually gets out, scenario by scenario.

We replay, with synthetic data and through your real egress controls, controlled exfiltration scenarios mapped to MITRE ATT&CK, and report for each what was attempted, transmitted and collected. Findings are then analysed with you to prioritise fixes.