Loading page...Загружаем страницу...正在加载页面...
AtlasRepo

I built Klarion, a secret scanner that asks an LLM before it alerts

说明与用途

Klarion 是一种在发出警报前询问 LLM 的秘密扫描器,减少了与 gitleaks 相比 33 倍、与 detect‑secrets 相比 1,000 倍以上的误报,同时保持了对真实秘密的高精度和召回率。

问题

Secret scanners generate many false positives, especially in typed languages, causing developers to ignore alerts and miss real leaks.

结果

Klarion reduced alerts by 33× compared to gitleaks and >1,000× compared to detect‑secrets, while achieving 1.7× the recall of gitleaks at similar precision.

工作流

  • Detect candidates using keyword gate, 81 regex rules, and normalized Rényi entropy.
  • For each candidate, an LLM reads the surrounding code and returns "secret", "false positive", or "uncertain" with a one‑line reason.
  • If the LLM flags a secret, Klarion raises an alert; otherwise it suppresses it.

证据

Klarion + model 11 alerts on the 4 repos, 0.18 per 1k files, 33x fewer alerts than gitleaks and over 1,000x fewer than detect‑secrets

来源

After it, Klarion's rate is 0.18 per 1k files on the typed repos and 0.20 on the control, so the language effect is gone.

来源

Klarion + model 0.35 recall, 0.89 precision; gitleaks 0.21 recall, 0.91 precision; detect‑secrets 0.35 recall, 0.34 precision.

来源

  • 记录日期: 2026-10-04T18:55:49.891Z