LLM in the Shell: Generative Honeypots
Open accessAbstract
Honeypots are essential tools in cybersecurity for early detection, threat intelligence gathering, and analysis of attacker's behavior. However, most of them lack the required realism to engage and fool human attackers long-term. Being easy to distinguish honeypots strongly hinders their effectiveness. This can happen because they are too deterministic, lack adaptability, or lack deepness. This work introduces shelLM, a dynamic and realistic software honeypot based on Large Language Models that generates Linux-like shell output. We designed and implemented shelLM using cloud-based LLMs. We evaluated if shelLM can generate output as expected from a real Linux shell. The evaluation was done by asking cybersecurity researchers to use the honeypot and give feedback if each answer from the honeypot was the expected one from a Linux shell. Results indicate that shelLM can create credible and dynamic answers capable of addressing the limitations of current honeypots. ShelLM reached a TNR of 0.90, convincing humans it was consistent with a real Linux shell. The source code and prompts for replicating the experiments have been publicly available.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In Sladic et al. (2024)'s experiment, most people cannot recognize whether the remote host is a real one or a honeypot generated by LLMs.
“In their experiment, most people can't recognize whether the remote host is a real one or a honeypot generated by LLMs.”
From When LLMs meet cybersecurity: a systematic literature review (2025) · cites this paper as Sladic et al. (2024)
Cites
Lost in the Middle: How Language Models Use Long Contexts
Liu, N. F., Lin, K., Hewitt, J., Paranjape, A., Bevilacqua, M., Petroni, F., & Liang, P. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00638 · Cited by 1,612 · cited 2× · see citing works
“False Discovery Rate = 𝐹𝑃 + 𝑇𝑃 (4)”
Background
Flagged first, then by citations
Cited by
Sorted by citations
A survey on large language model (LLM) security and privacy: The Good, The Bad, and The Ugly
Yao, Y., Duan, J., Xu, K., Cai, Y., Sun, Z., & Zhang, Y. (2024) · High-Confidence Computing · doi:10.1016/j.hcc.2024.100211 · Cited by 1,083 · see citing works
When LLMs meet cybersecurity: a systematic literature review
Zhang, J., Bu, H., Wen, H., Liu, Y., Fei, H., Xi, R., Li, L., Yang, Y., Zhu, H., & Meng, D. (2025) · Cybersecurity · doi:10.1186/s42400-025-00361-w · Cited by 220 · see citing works
“Specifically, the raw character sequences generated by the Codex are deserialized into an internal test case representation compatible with SBST to leverage its mutation operations and fitness functions.”
Generative AI in cybersecurity: A comprehensive review of LLM applications and vulnerabilities
Ferrag, M. A., Alwahedi, F., Battah, A., Cherif, B., Mechri, A., Tihanyi, N., Bisztray, T., & Debbah, M. (2025) · Internet of Things and Cyber-Physical Systems · doi:10.1016/j.iotcps.2025.01.001 · Cited by 148 · see citing works
Large Language Models in Cybersecurity: State-of-the-Art
Motlagh, F. N., Hajizadeh, M., Majd, M., Najafi, P., Cheng, F., & Meinel, C. (2024) · arXiv · doi:10.48550/arxiv.2402.00891 · Cited by 92 · see citing works
“In another study, Sladic et al. [79] present an LLM-based technique for developing software honeypots.”
MethodAdaptive Deception Architectures: Conceptual Foundations for LLM-Powered Honeypot Systems
Érsok, M., Balogh, Á., Kail, E., & Bánáti, A. (2025) · 2025 IEEE 19th International Symposium on Applied Computational Intelligence and Informatics (SACI) · doi:10.1109/saci66288.2025.11030110 · Cited by 2 · see citing works
- Authors
- Sladić, M., Valeros, V., Catania, C., & Garcia, S.
- Affiliations
- Czech Technical University in Prague (CZ)
- Consejo Nacional de Investigaciones Científicas y Técnicas (AR)
- Czech Technical University in Prague,Prague,Czech Republic
- Venue
- 2024 IEEE European Symposium on Security and Privacy Workshops (EuroS&PW)
- Publisher
- IEEE
- Published
- 2024-07-08
- Pages
- 430-435
- OpenAlex
- W4402157666
- Cited by
- 70
- References
- 1
- Citations by year
- 2026: 24 · 2025: 31 · 2024: 14 · 2023: 1
- FWCI
- 30.13
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE European Symposium on Security and Privacy Workshops (EuroS&PW) (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by-nc-nd)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Network Security and Intrusion Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Security and Verification in Computing
- Physical Sciences › Computer Science › Information Systems › Spam and Phishing Detection
- Keywords
- large language models, Linux shell, threat intelligence
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
