OpenBAS is Filigran’s open-source platform for testing whether an organisation’s defences and people actually hold up against realistic attacks. Since November 2025 the product has been called OpenAEV, short for Adversarial Exposure Validation.
It is built for security teams that have bought detection and response tools but cannot easily prove they work. Because the platform is driven by threat intelligence, the attacks it simulates reflect the threats a team is most likely to face.
At A Glance
| Vendor | Filigran |
|---|---|
| Category | Adversarial exposure validation, breach and attack simulation |
| Deployment | On-premises, SaaS hosted by Filigran, with or without an endpoint agent |
| Agentic AI | AI scenario generation, remediation guidance, XTM One agents |
| Customers | Swiss Federal Department of Foreign Affairs |
| Founded | 2022 (Filigran) |
| Market classification | Vulnerability Management |
What OpenBAS Does
Knowing that a control is installed is not the same as knowing it works. OpenAEV closes that gap by continuously firing real-world attack scenarios at the SIEM, EDR, firewall and SOC playbook controls a team relies on, then showing where detection or response broke down. People are tested as well as tools, because incident response depends on how staff and crisis teams behave under pressure. The results point teams to the defences and vulnerabilities that matter most for the threats they face, and to practical remediation, so fixes go where an attacker would actually get through.
Products
OpenAEV Community Edition
The open-source core, licensed under Apache 2.0 and deployed on-premises. Teams use it to plan, schedule and conduct adversary simulation campaigns and tabletop exercises, which suits organisations that want to start validating controls without a commercial agreement.
OpenAEV Enterprise Edition
Generally available since March 2026, the Enterprise Edition adds the automation larger organisations need and runs on-premises or as SaaS:
- Validation through EDR agents already on endpoints, so no extra agent is needed
- Attack chaining, which links individual actions and events into a full attack path
- A single, trackable exposure score for security posture
- The AI features described below
Human Readiness Exercises
Team-based simulations and social engineering scenarios test escalation paths and decision making before a real incident, so a crisis team finds its gaps in an exercise rather than during an attack.
How It Works
OpenAEV runs with or without an endpoint agent, across cloud, on-premises and multi-tenant setups. Self-managed installations use Docker. Filigran also runs a fully managed Enterprise instance with self-service provisioning.
The platform connects to existing tools through three kinds of connector. Injectors and executors carry out attack actions in the environment, and collectors bring the results back so each outcome can be scored. Threat intelligence flows in from Filigran’s OpenCTI platform, which shapes the scenarios. A test therefore runs end to end: a scenario is built from intelligence, executed through injectors or the EDR agents already on endpoints, and scored from what the collectors report about whether the activity was detected and stopped.
Agentic AI Capabilities
| Scenario generation | In the Enterprise Edition, AI processes threat intelligence and CERT reports and turns them into actions and new attack scenarios, so emerging threats can be tested soon after they are reported |
|---|---|
| Remediation recommendations | After a test, AI offers several remediation options with risk-based prioritisation and sequencing |
| XTM One | An AI orchestration layer, launched in June 2026, that connects OpenCTI and OpenAEV and supplies prepackaged agents for time-consuming tasks such as building attack scenarios |
Company Background
| Founded | 2022, by Samuel Hassine (CEO) |
|---|---|
| Headquarters | Paris, France |
| Funding | More than $100 million in total, including a $58 million Series C, with Eurazeo, Insight Partners, Accel and Deutsche Telekom’s T.Capital among the backers |
| Scale | Over 6,000 reported organisations worldwide |
Note: This overview is compiled from OpenBAS’s publicly available information as of 3 November 2026. It is provided for information only. NHI Mgmt Group does not endorse any vendor, and nothing on this page represents the opinions of NHI Mgmt Group.