mlsanityData to 7 Oct 2026

RepoAudit: An Autonomous LLM-Agent for Repository-Level Code Auditing

arXiv 2501.18160 ↗First posted 30 Jan 2025cs.SE cs.PLcited by 47

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

Code auditing is the process of reviewing code with the aim of identifying bugs. Large Language Models (LLMs) have demonstrated promising capabilities for this task without requiring compilation, while also supporting user-friendly customization. However, auditing a code repository with LLMs poses significant challenges: limited context windows and hallucinations can degrade the quality of bug reports, and analyzing large-scale repositories incurs substantial time and token costs, hindering efficiency and scalability. This work introduces an LLM-based agent, RepoAudit, designed to perform autonomous repository-level code auditing. Equipped with agent memory, RepoAudit explores the codebase on demand by analyzing data-flow facts along feasible program paths within individual functions. It further incorporates a validator module to mitigate hallucinations by verifying data-flow facts and checking the satisfiability of path conditions associated with potential bugs, thereby reducing false positives. RepoAudit detects 40 true bugs across 15 real-world benchmark projects with a precision of 78.43%, requiring on average only 0.44 hours and 2.54 per project. Also, it detects 185 new bugs in high-profile projects, among which 174 have been confirmed or fixed. We have open-sourced RepoAudit at https://github.com/PurCL/RepoAudit.

Key facts. Cited by 47 papers in the map (data through 7 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 19 of its 50 references that are in the map. First posted on arXiv 30 Jan 2025.

Builds on

19 of 50 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023ChatDev: Communicative Agents for Software Development16 Jul

Other references · 18

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2024The Llama 3 Herd of Modelscited by 14,192
  3. 2023Code Llama: Open Foundation Models for Codecited by 2,105
  4. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Societycited by 1,348
  5. 2024DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligencecited by 1,154
  6. 2024Qwen2.5-Coder Technical Reportcited by 1,151
  7. 2021CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generationcited by 890
  8. 2023YaRN: Efficient Context Window Extension of Large Language Modelscited by 584
  9. 2022Grounded Copilot: How Programmers Interact with Code-Generating Modelscited by 216
  10. 2024Vulnerability Detection with Code Language Models: How Far Are We?cited by 133
  11. 2023SatLM: Satisfiability-Aided Language Models Using Declarative Promptingcited by 100
  12. 2024IRIS: LLM-Assisted Static Analysis for Detecting Security Vulnerabilitiescited by 100
6 more
  1. 2023DiverseVul: A New Vulnerable Source Code Dataset for Deep Learning Based Vulnerability Detectioncited by 98
  2. 2023GPTScan: Detecting Logic Vulnerabilities in Smart Contracts by Combining GPT with Program Analysiscited by 65
  3. 2023Large Language Models for Code Analysis: Do LLMs Really Do Their Job?cited by 44
  4. 2024Top Score on the Wrong Exam: On Benchmarking in Machine Learning for Vulnerability Detectioncited by 27
  5. 2024Generative Multi-Modal Knowledge Retrieval with Large Language Modelscited by 22
  6. 2024LLMDFA: Analyzing Dataflow in Code with Large Language Modelscited by 22

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 16; 2026: 31.

47 papers.

Sort:

By area:

  1. 2025LLM-based Agents Suffer from Hallucinations: A Survey of Taxonomy, Methods, and Directionscited by 49
  2. 2026Graph-based Agent Memory: Taxonomy, Techniques, and Applicationscited by 22
  3. 2025Frontier AI's Impact on the Cybersecurity Landscapecited by 21
  4. 2025A Survey on Agentic Security: Applications, Threats and Defensescited by 17
  5. 2025VulnLLM-R: Specialized Reasoning LLM with Agent Scaffold for Vulnerability Detectioncited by 11
  6. 2026Detecting Safety Violations Across Many Agent Tracescited by 11
  7. 2025Validating Network Protocol Parsers with Traceable RFC Document Interpretationcited by 7
  8. 2025RFCAudit: An LLM Agent for Functional Bug Detection in Network Protocolscited by 7
  9. 2025Agentic Services Computingcited by 6
  10. 2026Co-RedTeam: Orchestrated Security Discovery and Exploitation with LLM Agentscited by 6
  11. 2025Large Language Models for Validating Network Protocol Parserscited by 5
  12. 2026Reducing False Positives in Static Bug Detection with LLMs: An Empirical Study in Industrycited by 5
35 more
  1. 2026AACR-Bench: Evaluating Automatic Code Review with Holistic Repository-Level Contextcited by 5
  2. 2025The Hitchhiker's Guide to Program Analysis, Part II: Deep Thoughts by LLMscited by 4
  3. 2025Automated Penetration Testing with LLM Agents and Classical Planningcited by 4
  4. 2026OpenSage: Self-programming Agent Generation Enginecited by 4
  5. 2026GBQA: A Game Benchmark for Evaluating LLMs as Quality Assurance Engineerscited by 4
  6. 2026AnyPoC: Universal Proof-of-Concept Test Generation for Scalable LLM-Based Bug Detectioncited by 4
  7. 2025Reasoning as a Resource: Optimizing Fast and Slow Thinking in Code Generation Modelscited by 3
  8. 2025textttReMind: Understanding Deductive Code Reasoning in LLMscited by 2
  9. 2025CoTDeceptor:Adversarial Code Obfuscation Against CoT-Enhanced LLM Code Agentscited by 2
  10. 2026Agentic AI-Empowered Dynamic Survey Frameworkcited by 2
  11. 2026Identifying Adversary Tactics and Techniques in Malware Binaries with an LLM Agentcited by 2
  12. 2026Automated Lemma Discovery in Agentic Program Verificationcited by 2
  13. 2026Antares: Foundation Models for Agentic Vulnerability Localizationcited by 2
  14. 2025Hound: Relation-First Knowledge Graphs for Complex-System Reasoning in Security Auditscited by 1
  15. 2026To Defend Against Cyber Attacks, We Must Teach AI Agents to Hackcited by 1
  16. 2026Argus: Reorchestrating Static Analysis via a Multi-Agent Ensemble for Full-Chain Security Vulnerability Detectioncited by 1
  17. 2026Veritas: Grounding LLM Agents for Reliable Vulnerability Reasoning over Stripped Binariescited by 1
  18. 2026Detecting Privilege Escalation in Polyglot Microservices via Agentic Program Analysiscited by 1
  19. 2026Revelio: Cost-Efficient Agentic Memory Safety Vulnerability Detection For Repository-Scale Codebasescited by 1
  20. 2026RepoRescue: An Empirical Study of LLM Agents on Whole-Repository Compatibility Rescuecited by 1
  21. 2026AutoTrace: From Patches to Triggers via Agentic Interprocedural Explorationcited by 1
  22. 2026EA-Graph: Artifact-Anchored Verification Memory for Coding Agents under Upstream Driftcited by 1
  23. 2025Position: Intelligent Coding Systems Should Write Programs with Justificationscited by 0
  24. 2025On the Effectiveness of Instruction-Tuning Local LLMs for Identifying Software Vulnerabilitiescited by 0
  25. 2026SPECA: Specification-to-Checklist Agentic Auditing for Multi-Implementation Systems -- A Case Study on Ethereum Clientscited by 0
  26. 2026Refute-or-Promote: An Adversarial Stage-Gated Multi-Agent Review Methodology for High-Precision LLM-Assisted Defect Discoverycited by 0
  27. 2026Beyond Pixels: Introspective and Interactive Grounding for Visualization Agentscited by 0
  28. 2026Agentic Repository Mining: A Multi-Task Evaluationcited by 0
  29. 2026Transferable Self-Evolving Playbooks for Agentic Security Auditingcited by 0
  30. 2026Evaluating LLMs for Real-World Web Vulnerability Detectioncited by 0
  31. 2026FlowArk: Boosting Agentic Data-flow Analysis for Android Apps via Context-Aware Knowledge Reusecited by 0
  32. 2026ALIBI: Adaptive Agentic Attacks on LLM-Based Vulnerability Detectors via Adversarial Code Commentscited by 0
  33. 2026OpenCodeReview: Determinism over Non-Determinism for Cost-Effective Agent-Based Code Reviewcited by 0
  34. 2026Plug 'n' Pray: Agentic LLM-based Detection of Potential Log File Exposures in Third-Party Content Management System Pluginscited by 0
  35. 2026WitnessGym: Benchmarking Coding Agents on the Construction of Bug Witnessescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026RepoZero: Can LLMs Generate a Code Repository from Scratch?
  2. 2026E2E-SWE: Benchmarking LLMs on Building Working Codebases from Scratch
  3. 2024CodeAgent: Enhancing Code Generation with Tool-Integrated Agent Systems for Real-World Repo-level Coding Challenges
  4. 2026RepoLaunch: Automating Build and Management of Code Repositories across Languages and Platforms
  5. 2026Large-scale Repository Engineering via Agent-Native Reusable Code Primitives
  6. 2026RepoMod-Bench: A Benchmark for Code Repository Modernization via Implementation-Agnostic Testing
  7. 2026AgentSZZ: Teaching the LLM Agent to Play Detective with Bug-Inducing Commits
  8. 2025Benchmarking LLMs and LLM-based Agents in Practical Vulnerability Detection for Code Repositories