Welcome to CredencePlus
A Benchmarking Tool for your LLM's Threat Intelligence
LLM Assistant
Evaluation Results
| Model | MCQ Accuracy | RCM Accuracy | VSP MAD | TAA Accuracy |
|---|
Performance Comparison
Qualitative âSkills & Gapsâ Report
Strengths
You scored 92% on MCQ, which shows your team has solid foundational CTI knowledge of mitigations and attacker tactics.
Weaknesses
Your CVSSâvector predictions averaged 58% accuracyâsuggesting you may be underestimating the impact of complex attack chains.
Recommendations
MCQ: Continue hands-on red-team drills on the newest CWE classes (e.g. supply-chain, cloud-native).
VSP: Adopt an automated CVSS-calculator plugin or retrain your scoring guidelines around UI/PR distinctions.
RCM: Introduce a peer-review step on ambiguous CVEâCWE mappings and integrate richer code examples into your playbooks.
TAA: Map your real incident data against MITRE ATT&CK to close gaps.
Trusted by Industry Leaders
Google Sec - Gemini v1
Cisco
Trend Cybertron