BlackIce (Databricks)
Containerized red teaming toolkit for AI security testing.
What it does
A containerised red teaming toolkit from Databricks designed for AI security testing. Ships as Docker containers with pre-configured testing environments and attack tooling.
Security relevance
The containerised approach ensures reproducible test environments — critical when you need to demonstrate consistent findings across engagements. Integrates with Databricks' ML ecosystem but can be used standalone against any LLM endpoint.
When to use it
Use when you need isolated, reproducible red team environments, particularly in regulated industries where test environment consistency matters. Requires Docker knowledge and red team experience.
OWASP coverage
Risks addressed — mapped to both OWASP Top 10 standards. 3 in LLM, 2 in Agentic.
The raw record
What Yuntona stores. Single source of truth — fork it on GitHub.
name: BlackIce (Databricks) slug: blackice-databricks type: Mixed category: AI Red Teaming url: https://www.databricks.com/blog/announcing-blackice-containerized-red-teaming-toolkit-ai-security-testing reviewed: 2026-04 added: 2026-04 updated: 2026-04 risks: llm: [LLM01, LLM02, LLM06] asi: [ASI01, ASI02] complexity: Expert Required pricing: — audience: Red Team lifecycle: [test] tags: [Container, Pentesting, Tool]