almanax/w3sa-bm-stellar

W3SA - Stellar Codebase Benchmark

Overview

This repository contains the code for the W3SA Benchmark for Stellar.

Repo Structure

The benchmark contains two folders, a benchmark and an src folder. The benchmark folder has all the projects use for eval along with their audit findings in the ground_truth and the src folder contains the scripts used to generate these outputs, allowing for reproducibility and further analysis.

├── README.md
├── benchmark
│   ├── config/
│   ├── ground_truth/
│   └── repositories/
└── bm_src
    ├── dataset_transformation.py
    ├── eval.py
    ├── experiments.py
    ├── models.py
    ├── prompts.py
    └── metrics.py
 

Project Statistics

Project details with total number of vulnerabilities for each severity level

Project Name	Critical/High	Medium	Low/ Informational
reflector	1	1	4
slender	12	9	7
soroswap	1	5	4
comet	2	4	9
allbridge-core	0	1	8
Total	16	20	32

Detection Rate

Projects	Claude 3.5	4o	o3-mini	o1-mini	o1	ALMX
reflector	0.33	0.33	0.33	0.16	0.5	0.5
slender	0.28	0.07	0.14	0.07	0.14	0.25
soroswap	0.1	0.1	0.1	0.2	0.3	0.3
comet	0.06	0.06	0.13	0.06	0.13	0.33
allbridge-core	0.22	0.22	0.22	0.22	0.11	0.33
Total	0.198	0.156	0.184	0.142	0.236	0.342

Set up

• Install uv package manager if not yet available
• Run uv sync

Run an experiment

• Set your OPENAI_API_KEY as environmental variable
• Launch your experiment by running:

uv run experiment.py --model o3-mini

Contact Us

For or questions, suggestions, or to learn more about Almanax.ai, reach out to us at https://www.almanax.ai/contact