Introducing SOLVE+

March 2, 2026

In this article

    March 2, 2026

    Last year, we published SOLVE, a scoring framework for assessing the difficulty of vulnerability research and exploit development challenges. However, offensive cyber operations require a broader set of capabilities; cybersecurity operators must gather intelligence, maintain stealth, manipulate human targets, and coordinate multi-phase campaigns. Our CyScenarioBench evaluation framework was designed to test these broader capabilities, extending beyond vulnerability research and exploit development challenges.

    SOLVE+ extends the SOLVE scoring framework to cover these additional capabilities, allowing scoring of the more expansive challenges in CyScenarioBench. Specifically, SOLVE+ extends SOLVE's vulnerability research and exploitation scoring with four additional capability areas:

    • Intelligence Gathering & Reconnaissance - finding, correlating, and operationalizing information across external and internal sources.

    • Operational Security - understanding defensive environments, applying evasion and anti-forensic techniques, adapting in real time, and maintaining operational discipline.

    • Social Engineering - researching human targets, crafting convincing pretexts, and deploying credible supporting infrastructure.

    • Planning & Orchestration - managing operation scope, dependencies, and integrating diverse tools and skill sets.

    Each capability is broken into sub-capabilities, with defined dimensions of difficulty, scored 0–10 based on what the challenge demands.

    Because these capabilities are fundamentally different in nature, challenges with similar overall scores can vary significantly in what they actually require. We use the same maxplus aggregation as the original SOLVE to produce a single score; however, this score may be misleading: We generally recommend looking at individual capability scores, which are more informative for understanding where the difficulty lies in different challenges.

    This is a very early version of the framework. The capability areas and scoring criteria are likely to evolve as we refine and field-test the system. This early version is shared with a select group of partners for feedback ahead of a broader public release.

    Read the full SOLVE+ framework here.

    March 2, 2026

    Last year, we published SOLVE, a scoring framework for assessing the difficulty of vulnerability research and exploit development challenges. However, offensive cyber operations require a broader set of capabilities; cybersecurity operators must gather intelligence, maintain stealth, manipulate human targets, and coordinate multi-phase campaigns. Our CyScenarioBench evaluation framework was designed to test these broader capabilities, extending beyond vulnerability research and exploit development challenges.

    SOLVE+ extends the SOLVE scoring framework to cover these additional capabilities, allowing scoring of the more expansive challenges in CyScenarioBench. Specifically, SOLVE+ extends SOLVE's vulnerability research and exploitation scoring with four additional capability areas:

    • Intelligence Gathering & Reconnaissance - finding, correlating, and operationalizing information across external and internal sources.

    • Operational Security - understanding defensive environments, applying evasion and anti-forensic techniques, adapting in real time, and maintaining operational discipline.

    • Social Engineering - researching human targets, crafting convincing pretexts, and deploying credible supporting infrastructure.

    • Planning & Orchestration - managing operation scope, dependencies, and integrating diverse tools and skill sets.

    Each capability is broken into sub-capabilities, with defined dimensions of difficulty, scored 0–10 based on what the challenge demands.

    Because these capabilities are fundamentally different in nature, challenges with similar overall scores can vary significantly in what they actually require. We use the same maxplus aggregation as the original SOLVE to produce a single score; however, this score may be misleading: We generally recommend looking at individual capability scores, which are more informative for understanding where the difficulty lies in different challenges.

    This is a very early version of the framework. The capability areas and scoring criteria are likely to evolve as we refine and field-test the system. This early version is shared with a select group of partners for feedback ahead of a broader public release.

    Read the full SOLVE+ framework here.

    BACK TO PUBLICATIONS