HN Showcase
Back to browse
SecLens – Claude Opus scores 0.000 on auth vulns, 0.689 on SSRF

SecLens – Claude Opus scores 0.000 on auth vulns, 0.689 on SSRF

by subho007·Apr 1, 2026·4 points·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidDark HorseBig BrainEye Candy

Stakeholder-weighted LLM security benchmark reveals 31-point score swings for the same model.

Strengths
  • •Role-based weighting exposes hidden model weaknesses aggregate scores miss.
  • •406 CVE tasks across 12 models provides substantial empirical data.
  • •Clean interactive leaderboard makes complex benchmark data immediately digestible.
Weaknesses
  • •Benchmark projects are reference material, not tools with recurring utility.
  • •No clear path to integrate weights into actual model selection workflows.
Category
AI/ML
Target Audience

Security teams and CTOs evaluating LLMs for security tasks

Similar To

SecurityBench · Hugging Face Open LLM Leaderboard · LiveBench

Similar Projects