Skip to content
CounterICE

Products

Three products, one standard.

The data comes first. Everything we build is held to its graders' rule: a result counts only if it can be re-checked.

Available now

Product: -[CONTAIN]-Post-train RL Dataset Batch 1

Security tasks written and reviewed by working researchers, covering exploitation, review, and triage across Web2, Web3, and the software supply chain, each a live, auto-graded environment.

-[CONTAIN]-
Tasks
3,000
Domains
Frontend injection / Blockchain attack / Supplychain attack / Algorithms
Formats
Harbor / Anthropic
Explore sample dataset
Early access

Product: -[SECURE]-Agentic Security Validation Swarm

Give your team evidence to act on. Coordinated agents scan web apps and smart contracts, investigate attack paths, and help validate fixes.

-[SECURE]-
Coverage
Web apps · EVM · Solana preview
Method
Scan · investigate · validate · retest
Output
Findings · replay evidence · reports
In partnership with
Explore the workbench

CounterICE

Secure. Contain. Protect.