# Multi-Agent Taint Specification Extraction for Vulnerability Detection

arXiv: https://arxiv.org/abs/2601.10865 · HTML: https://mlsanity.com/paper/2601.10865/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 22 of its 86 references that are in the map.
- First posted on arXiv 15 Jan 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)

## Builds on (most cited first, 22 of 22)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMs](https://mlsanity.com/paper/2306.13063/) (2023)
- [ChatDev: Communicative Agents for Software Development](https://mlsanity.com/paper/2307.07924/) (2023)
- [IRIS: LLM-Assisted Static Analysis for Detecting Security Vulnerabilities](https://mlsanity.com/paper/2405.17238/) (2024)
- [Non-Determinism of "Deterministic" LLM Settings](https://mlsanity.com/paper/2408.04667/) (2024)
- [Small World with High Risks: A Study of Security Threats in the npm Ecosystem](https://mlsanity.com/paper/1902.09217/) (2019)
- [From CVE Entries to Verifiable Exploits: An Automated Multi-Agent Framework for Reproducing CVEs](https://mlsanity.com/paper/2509.01835/) (2025)
- [PyCG: Practical Call Graph Generation in Python](https://mlsanity.com/paper/2103.00587/) (2021)
- [A Simple Ensemble Strategy for LLM Inference: Towards More Stable Text Classification](https://mlsanity.com/paper/2504.18884/) (2025)
- [QLPro: Automated Code Vulnerability Discovery via LLM and Static Code Analysis Integration](https://mlsanity.com/paper/2506.23644/) (2025)
- [Utilizing Precise and Complete Code Context to Guide LLM in Automatic False Positive Mitigation](https://mlsanity.com/paper/2411.03079/) (2024)
- [Open Source, Open Threats? Investigating Security Challenges in Open-Source Software](https://mlsanity.com/paper/2506.12995/) (2025)
- [Study of JavaScript Static Analysis Tools for Vulnerability Detection in Node.js Packages](https://mlsanity.com/paper/2301.05097/) (2023)
- [Seneca: Taint-Based Call Graph Construction for Java Object Deserialization](https://mlsanity.com/paper/2311.00943/) (2023)
- [Identifying Helpful Context for LLM-based Vulnerability Repair: A Preliminary Study](https://mlsanity.com/paper/2506.11561/) (2025)
- [LLM-Driven SAST-Genius: A Hybrid Static Analysis Framework for Comprehensive and Actionable Security](https://mlsanity.com/paper/2509.15433/) (2025)
- [Accelerating JavaScript Static Analysis via Dynamic Shortcuts (Extended Version)](https://mlsanity.com/paper/2105.13699/) (2021)
- [InspectJS: Leveraging Code Similarity and User-Feedback for Effective Taint Specification Inference for JavaScript](https://mlsanity.com/paper/2111.09625/) (2021)
- [npm-follower: A Complete Dataset Tracking the NPM Ecosystem](https://mlsanity.com/paper/2308.12545/) (2023)
- [Call Me Maybe: Enhancing JavaScript Call Graph Construction using Graph Neural Networks](https://mlsanity.com/paper/2506.18191/) (2025)

## Built on by (most cited first, 3 of 3)

- [Understanding Implicit Trust Errors in Core Carrier Networks through Multi-Agent Flaw Discovery and Analysis](https://mlsanity.com/paper/2607.10315/) (2026)
- [FlowArk: Boosting Agentic Data-flow Analysis for Android Apps via Context-Aware Knowledge Reuse](https://mlsanity.com/paper/2607.11308/) (2026)
- [The History Is the Detector: Executing CVE Patch History, End-to-End](https://mlsanity.com/paper/2609.05335/) (2026)

## Abstract (arXiv)

Static Application Security Testing (SAST) tools using taint analysis are widely viewed as providing higher-quality vulnerability detection results compared to traditional pattern-based approaches. However, performing static taint analysis for JavaScript poses two major challenges. First, JavaScript's dynamic features complicate data flow extraction required for taint tracking. Second, npm's large library ecosystem makes it difficult to identify relevant sources/sinks and establish taint propagation across dependencies. In this paper, we present SemTaint, a multi-agent system that strategically combines the semantic understanding of Large Language Models (LLMs) with traditional static program analysis to extract taint specifications, including sources, sinks, call edges, and library flow summaries tailored to each package. Conceptually, SemTaint uses static program analysis to calculate a call graph and defers to an LLM to resolve call edges that cannot be resolved statically. Further, it uses the LLM to classify sources and sinks for a given CWE. The resulting taint specification is then provided to a SAST tool, which performs vulnerability analysis. We integrate SemTaint with CodeQL, a state-of-the-art SAST tool, and demonstrate its effectiveness by detecting 106 of 162 vulnerabilities previously undetectable by CodeQL. Furthermore, we find 4 novel vulnerabilities in 4 popular npm packages. In doing so, we demonstrate that LLMs can practically enhance existing static program analysis algorithms, combining the strengths of both symbolic reasoning and semantic understanding for improved vulnerability detection.