From 73ba98afa8f13680c8495fae8e1d90da6f019ebb Mon Sep 17 00:00:00 2001 From: tduhamel42 Date: Thu, 16 Oct 2025 14:05:05 +0200 Subject: [PATCH] docs: Add Secret Detection Benchmarks section with performance metrics - Added dedicated section showcasing secret detection benchmark results - Includes comparison table with recall rates and speeds - Links to detailed benchmark analysis - Highlights LLM detector's 84.4% recall on obfuscated secrets --- README.md | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/README.md b/README.md index 7af590c..b88eeff 100644 --- a/README.md +++ b/README.md @@ -67,6 +67,23 @@ If you find FuzzForge useful, please star the repo to support development 🚀 --- +## 🔍 Secret Detection Benchmarks + +FuzzForge includes three secret detection workflows benchmarked on a controlled dataset of **32 documented secrets** (12 Easy, 10 Medium, 10 Hard): + +| Tool | Recall | Secrets Found | Speed | +|------|--------|---------------|-------| +| **LLM (gpt-5-mini)** | **84.4%** | 41 | 618s | +| **LLM (gpt-4o-mini)** | 56.2% | 30 | 297s | +| **Gitleaks** | 37.5% | 12 | 5s | +| **TruffleHog** | 0.0% | 1 | 5s | + +📊 [Full benchmark results and analysis](backend/benchmarks/by_category/secret_detection/results/comparison_report.md) + +The LLM-based detector excels at finding obfuscated and hidden secrets through semantic analysis, while pattern-based tools (Gitleaks) offer speed for standard secret formats. + +--- + ## 📦 Installation ### Requirements