Projects sharing features with Seclens

This is the codebase for the paper "Generalization-Enhanced Code Vulnerability Detection via Multi-Task Instruction Fine-Tuning", a research outcome by the ARTS^3 research group from Huazhong University of Science and Technology.

LLMDFA is an LLM-powered data-flow analysis framework. Specifically, it instantiates bottom-up summary-based data-flow analysis by interpreting intra-procedural data-flow facts with LLMs. With the specified sources/sinks and data-flow transfer functions (i.e., rules of propagating data-flow…

LLMSAN is a tool for prompting-based bug detection. Equipped with the sanitization technique, LLMSAN can recognize false positives in the reported bugs without introducing huge additional token costs.

📜 Overview | 📚 Dataset | 💻 Experiments | 📝 Citation

PurpleLlama is a collection of security toolsets and frameworks designed to audit large language model vulnerabilities and implement runtime input-output guardrails. It provides a security evaluation framework and benchmark suite to quantify risks associated with prompt injections and the generation of malicious code. The project includes a content moderator and input-output filters that use a standardized taxonomy to identify and block harmful content, jailbreaking attempts, and insecure commands. It also features capabilities for sensitive document classification to prevent the unauthorized

Code Vulnerability Detection with Devign and RV Datasets This repository contains code for training and evaluating a vulnerability detection model using the Devign and RV datasets. The model is based on RoBERTa and UnixCoder, leveraging pre-trained models for code defect prediction tasks. This…

This repository contains materials for the paper "How Far Have We Gone in Vulnerability Detection Using Large Language Model".

SV‑TrustEval‑C is the first reasoning‑based benchmark designed to rigorously evaluate Large Language Models (LLMs) on both structure (control/data flow) and semantic reasoning for vulnerability analysis in C source code. Unlike existing benchmarks that focus on pattern recognition,…

Shannon is an integrated security platform designed for autonomous penetration testing, static and dynamic analysis, and automated vulnerability remediation within self-hosted, private infrastructure. It functions as a unified security suite that orchestrates the entire lifecycle of vulnerability management, from initial discovery and reachability prioritization to the generation and verification of code-level patches. The platform distinguishes itself through its agentic approach to security, deploying autonomous agents to execute both black-box and white-box exploits against running applica

LLM-SmartAudit is a cutting-edge tool designed to revolutionize smart contract auditing using advanced language models.

This is the replication package accompanying our paper "R2Vul: Learning to Reason about Software Vulnerabilities with Reinforcement Learning and Structured Reasoning Distillation".

Garak is an AI model evaluation tool and vulnerability scanner designed for red teaming large language models and auditing the security of retrieval-augmented generation pipelines. It identifies behavioral weaknesses, such as jailbreaks, hallucinations, and data leakage, by simulating adversarial attacks and executing automated testing vectors. The framework utilizes an adaptive probing loop where prompts can react to previous model behavior and be modified in flight via middleware. To ensure consistent analysis, it employs a provider-agnostic interface to interact with various model APIs and

Promptfoo is an evaluation framework designed for testing, benchmarking, and red-teaming language models and agentic workflows. It provides a unified environment to run prompts against multiple providers, allowing developers to systematically validate model outputs against objective assertions, semantic similarity metrics, and custom grading rubrics. The platform distinguishes itself through a provider-agnostic execution layer and a stateful orchestrator capable of simulating multi-turn conversations and complex tool-use trajectories. It includes a dedicated adversarial mutation pipeline that

Machine Learning (ML) models are shared publicly over the internet, within teams and across teams. The rise of Foundation Models have resulted in public ML models being increasingly consumed for further training/fine tuning. ML Models are increasingly used to make critical decisions and power…

2025.11 LLMxCPG wins the MENA CSAW Applied Research Competition (ARC) 2025. - 2025.10 LLMxCPG is accepted to the MENA CSAW Applied Research Competition (ARC) 2025. - 2025.08 LLMxCPG is presented at USENIX Security 2025.

This repository contains the necessary code for reproducing the experiments conducted in the paper "Finetuning Large Language Models for Vulnerability Detection": https://doi.org/10.1109/ACCESS.2025.3546700

The complete record of the discussion example introduced in the paper "Multi-role Consensus through LLMs Discussions for Vulnerability Detection".

This is the codebase for the paper "Investigating Large Language Models for Code Vulnerability Detection: An Experimental Study", which is submitted to IEEE TIFS. The arxiv version of this work is publicly available via this link. We also provide the 🤗Huggingface repository for all evaluated…

This repo contains all the code and test cases for the LLift project

VulDetectBench is a benchmark designed to evaluate the vulnerability detection capabilities of Large Language Models (LLMs).

VulnGym: A Real-World, Project-Level Vulnerability Benchmark for White-Box Vulnerability-Hunting Agents

Strix is an automated security research and vulnerability scanning platform that leverages language models to orchestrate complex security analysis tasks. It functions as a comprehensive framework for penetration testing and continuous security integration, allowing users to embed automated vulnerability research directly into development pipelines or execute it within isolated, containerized environments. The platform distinguishes itself through a multi-agent orchestration engine that coordinates specialized autonomous agents to perform parallel security assessments. By integrating LLM-agno

Software vulnerabilities are the weaknesses in software systems, that leads to serious cybersecurity problems. Recently, many deep learning-based approaches have been proposed to detect vulnerabilities at the function level by using one or a few different modalities (e.g., text representation,…

An intelligent vulnerability detection framework based on Large Language Models and static code analysis, supporting multi-dataset evaluation and ablation studies.