Poisoned Skills：越狱攻击可通过 Agent 的 Markdown 技能文件传播

View PDF HTML (experimental)

Abstract:Modern agent frameworks allow users to provide Markdown files containing system prompts, tool descriptions, and skill definitions (e.g., this http URL), which are incorporated into the agent's context during initialization. In this work, we investigate whether this configuration mechanism can also serve as an injection channel: can a jailbreak embedded in a Markdown skill file propagate through an agent and induce prohibited behavior? To answer this question, we construct a LangChain-based question-answering agent whose system prompt is assembled from a user-provided Markdown file. We embed a jailbreak combining persona manipulation with a GPTFUZZER-based attack and evaluate its effectiveness across seven contemporary open-weight models using 270 forbidden questions spanning nine harm categories. Under a benign system prompt, the models exhibit near-complete refusal behavior, with a mean attack success rate (ASR) of just 0.5%. In contrast, the injected skill file raises the mean ASR to 45%, reaching 68.9% for individual models. We further evaluate two payload obfuscation strategies: an invisible zero-width character-based stealth payload and a phonetic payload based on grapheme-to-phoneme transformations. Neither outperforms the plain-text injection; notably, the phonetic variant reduces mean ASR to 30%. These findings identify Markdown skill files as an underexplored attack surface for AI agents and suggest that vulnerability to configuration-based jailbreaks depends more strongly on the underlying model than on the specific payload obfuscation technique.

Subjects:	Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
Cite as:	arXiv:2610.37738 [cs.CR]
	(or arXiv:2610.37738v1 [cs.CR] for this version)
	https://doi.org/10.48550/arXiv.2610.37738 arXiv-issued DOI via DataCite

Submission history

From: Murali Emani [view email]
[v1] Tue, 27 Oct 2026 15:29:41 UTC (32 KB)