EPSS (Exploit Prediction Scoring System)
The Exploit Prediction Scoring System (EPSS) is a data-driven framework designed to estimate the likelihood that a software vulnerability will be exploited in the wild. It helps security professionals prioritize which vulnerabilities to address first based on real-world risk, rather than just severity.
What EPSS Measures
EPSS provides a probability score (0 to 1) indicating how likely it is that a vulnerability will be exploited within a short time frame (typically the next 30 days). For example:
- EPSS Score of 0.6 means there's a 60% chance of exploitation.
- EPSS Score of 0.01 means there's only a 1% chance.
How EPSS Works
EPSS uses machine learning models trained on:
- CVE metadata (e.g., CVSS scores, affected software)
- Exploit availability (e.g., public exploit code)
- Threat intelligence feeds
- Historical exploitation data
This allows EPSS to dynamically assess risk based on current trends and attacker behavior.
Why EPSS Is Useful
- Prioritization: Helps focus remediation efforts on vulnerabilities most likely to be exploited.
- Complement to CVSS: CVSS measures severity, but not exploit likelihood. EPSS fills that gap.
- Real-world relevance: Based on actual exploitation data, not theoretical risk.
EPSS vs CVSS
Use Cases
- Vulnerability management: Prioritize patching based on EPSS scores.
- Risk assessment: Combine EPSS with asset value and exposure.
- Threat modeling: Identify high-risk vulnerabilities in attack paths.