When "Correct" Is Not Safe: Can We Trust Functionally Correct Patches Generated by Code Agents?
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912662422552576 |
|---|---|
| author | Peng, Yibo Song, James Li, Lei Yang, Xinyu Christodorescu, Mihai Mangal, Ravi Pasareanu, Corina Zheng, Haizhong Chen, Beidi |
| author_facet | Peng, Yibo Song, James Li, Lei Yang, Xinyu Christodorescu, Mihai Mangal, Ravi Pasareanu, Corina Zheng, Haizhong Chen, Beidi |
| contents | Code agents are increasingly trusted to autonomously fix bugs on platforms such as GitHub, yet their security evaluation focuses almost exclusively on functional correctness. In this paper, we reveal a novel type of threat to real-world code agents: Functionally Correct yet Vulnerable (FCV) patches, which pass all test cases but contain vulnerable code. With our proposed FCV-Attack, which can be deliberately crafted by malicious attackers or implicitly introduced by benign developers, we show that SOTA LLMs (e.g., ChatGPT and Claude) and agent scaffolds (e.g., SWE-agent and OpenHands) are all vulnerable to this FCV threat; across 12 agent-model combinations on SWE-Bench, the attack only requires black-box access and a single query to the code agent to perform the attack. For example, for CWE-538 (information exposure vulnerability), the FCV-Attack attains an attack success rate of $40.7\%$ on GPT-5 Mini + OpenHands. Our results reveal an important security threat overlooked by current evaluation paradigms and urge the development of security-aware defenses for code agents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_17862 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | When "Correct" Is Not Safe: Can We Trust Functionally Correct Patches Generated by Code Agents? Peng, Yibo Song, James Li, Lei Yang, Xinyu Christodorescu, Mihai Mangal, Ravi Pasareanu, Corina Zheng, Haizhong Chen, Beidi Cryptography and Security Software Engineering Code agents are increasingly trusted to autonomously fix bugs on platforms such as GitHub, yet their security evaluation focuses almost exclusively on functional correctness. In this paper, we reveal a novel type of threat to real-world code agents: Functionally Correct yet Vulnerable (FCV) patches, which pass all test cases but contain vulnerable code. With our proposed FCV-Attack, which can be deliberately crafted by malicious attackers or implicitly introduced by benign developers, we show that SOTA LLMs (e.g., ChatGPT and Claude) and agent scaffolds (e.g., SWE-agent and OpenHands) are all vulnerable to this FCV threat; across 12 agent-model combinations on SWE-Bench, the attack only requires black-box access and a single query to the code agent to perform the attack. For example, for CWE-538 (information exposure vulnerability), the FCV-Attack attains an attack success rate of $40.7\%$ on GPT-5 Mini + OpenHands. Our results reveal an important security threat overlooked by current evaluation paradigms and urge the development of security-aware defenses for code agents. |
| title | When "Correct" Is Not Safe: Can We Trust Functionally Correct Patches Generated by Code Agents? |
| topic | Cryptography and Security Software Engineering |
| url | https://arxiv.org/abs/2510.17862 |