2606.09084v1 Jun 08, 2026 cs.CR

Context-Fractured Decomposition Attacks on Tool-Using LLM Agents: Exploiting Artifact Provenance Gaps

Xiaofeng Lin
Xiaofeng Lin
Citations: 76
h-index: 4
Guang Cheng
Guang Cheng
Citations: 16
h-index: 2
Charles Fleming
Charles Fleming
Citations: 23
h-index: 3
Yukai Yang
Yukai Yang
Citations: 0
h-index: 0
Daniel Guo
Daniel Guo
Citations: 1,117
h-index: 3
Sahil Arun Nale
Sahil Arun Nale
Citations: 0
h-index: 0

Tool-using LLM agents interact with the world through actions that persist state in artifacts (e.g., workspace files or logs). Consequently, jailbreak defenses must reason about cross-step composition rather than isolated text. Yet most existing attacks and defenses, including ``multi-turn'' jailbreaks such as Crescendo and Tree of Attacks,still assume a single contiguous conversation visible to the defender. This assumption breaks down in real agent pipelines, where enforcement is fragmented across tools, modules, and time, and where artifact provenance is often not tracked. We operationalize a deployment failure mode for tool-using LLM agents, the \emph{provenance gap}, and study reproducible triggers for it: \emph{Context-Fractured Decomposition} (CFD), a family of cross-context multi-step jailbreaks that preserve benign-looking intermediate artifacts from an early interaction and elicit harmful behavior much later, potentially in a different agent instance or workflow stage, via individually innocuous tool actions whose risk emerges only under delayed artifact-mediated composition. We instrument the failure mode with trace-level diagnostics and outline a verifiable mitigation direction (provenance lineage tagging). Across agent-system jailbreak benchmarks, CFD improves success rates by up to 28.3 percentage points over state-of-the-art baselines, even against strong single-turn judges. Disclaimer: This paper contains examples of harmful or offensive language.

0 Citations
0 Influential
2 Altmetric
10.0 Score
Original PDF

No Analysis Report Yet

This paper hasn't been analyzed by Gemini yet.

Log in to request an AI analysis.

댓글

댓글을 작성하려면 로그인하세요.

아직 댓글이 없습니다. 첫 번째 댓글을 남겨보세요!