Towards a Risk Assessment of Malicious Skill Files in Coding Agents

A new study from arXiv (submitted Aug 5, 2026) finds that coding agents are highly vulnerable to malicious skill files, with Gemini CLI exploited in 95.5-96.1% of runs and Qwen Code in 71.6-74.0% of runs across 5,629 completed runs. The researchers generated 2,826 adversarial skills from 471 real-world shell commands using six LLMs across four families, mapped to 11 MITRE ATT&CK tactics, and found explicit safety recognition in only 1.99% of runs, urging enterprises to assess skill-interface risk before adoption.

Computer Science Software Engineering Submitted on 5 Aug 2026 Title:Towards a Risk Assessment of Malicious Skill Files in Coding Agents View PDF /pdf/2608.05223 HTML experimental https://arxiv.org/html/2608.05223v1 Abstract:Autonomous coding agents are increasingly embedded in enterprise software workflows with delegated authority over connected systems. Central to this architecture is the agent skills interface: folders of instructions and scripts that agents load dynamically to specialize their behavior. This interface also widens the attack surface, letting malicious shell commands hide within natural-language skill files. We make three contributions. First, an adversarial skill-synthesis method using six LLMs across four families to transform 471 real-world shell commands into benign-appearing skills, released as a benchmark of 2,826 skills mapped to 11 MITRE ATT&CK tactics. Second, a reproducible evaluation pipeline coupling run stratification, evidence anchoring, a refusal veto, and a deterministic declared-intent override with a three-judge LLM-as-a-judge panel, validated against a blind human gold standard Cohen's kappa = 0.85 . Third, a large-scale characterization of two enterprise-grade agents across 5,629 completed runs. Gemini CLI is exploited in 95.5-96.1% of runs and Qwen Code in 71.6-74.0% raw majority vote to declared-intent-corrected estimate, both within the human gold standard , nearly invariant to the generating model. Explicit safety recognition occurs in only 1.99% of runs. Enterprises must assess and mitigate skill-interface risk before adopting coding agents. Our code and dataset are available at this https URL References & Citations Loading... Bibliographic and Citation Tools Bibliographic Explorer What is the Explorer? https://info.arxiv.org/labs/showcase.html arxiv-bibliographic-explorer Connected Papers What is Connected Papers? https://www.connectedpapers.com/about Litmaps What is Litmaps? https://www.litmaps.co/ scite Smart Citations What are Smart Citations? https://www.scite.ai/ Code, Data and Media Associated with this Article alphaXiv What is alphaXiv? https://alphaxiv.org/ CatalyzeX Code Finder for Papers What is CatalyzeX? https://www.catalyzex.com DagsHub What is DagsHub? https://dagshub.com/ Gotit.pub What is GotitPub? http://gotit.pub/faq Hugging Face What is Huggingface? https://huggingface.co/huggingface ScienceCast What is ScienceCast? https://sciencecast.org/welcome Demos Recommenders and Search Tools Influence Flower What are Influence Flowers? https://influencemap.cmlab.dev/ CORE Recommender What is CORE? https://core.ac.uk/services/recommender arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs https://info.arxiv.org/labs/index.html .