Package Hallucination Attacks on Coding Agents through Prompt Injection in Rule Files

arXiv:2610.09264 · cs.CR, cs.AI · Submitted 2026-10-07 · Read on arXiv

cs.CR, cs.AI

Submitted: 2026-10-07

Updated: 2026-10-07

Code: https://github.com/agentsmd/agents.md

Terminology

• Hallucination — 51× in this paper · explained in 3 episodes
• Benchmark — 8× in this paper · explained in 2 episodes
• Transferability — 8× in this paper · explained in 4 episodes
• Agentic Coding — 7× in this paper · explained in 1 episode
• Adam — 4× in this paper · explained in 2 episodes
• Trace — 4× in this paper · explained in 2 episodes
• Ablation — 2× in this paper · explained in 1 episode
• Benchmarking — 2× in this paper · explained in 3 episodes
• Deployable Attack Success Rate — 2× in this paper
• Generalization — 2× in this paper · explained in 13 episodes
• Syntactic Attack Success Rate — 2× in this paper
• Ablation Studies — 1× in this paper · explained in 1 episode
• Ablation study — 1× in this paper · explained in 1 episode
• Adversarial Attacks — 1× in this paper · explained in 1 episode
• Agentic AI — 1× in this paper · explained in 1 episode
• Core — 1× in this paper · explained in 2 episodes
• Greedy Coordinate Gradient — 1× in this paper
• Loss Function — 1× in this paper · explained in 2 episodes
• Multi-Agent System — 1× in this paper · explained in 6 episodes
• Quantization — 1× in this paper · explained in 12 episodes
• Reasoning Traces — 1× in this paper · explained in 3 episodes
• Sparsity — 1× in this paper · explained in 4 episodes
• Verification — 1× in this paper · explained in 2 episodes

Sources

• Universal and Transferable Adversarial Attacks on Aligned Language Models
• OpenAI GPT-5 System Card
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• Large Language Models Hallucination: A Comprehensive Survey
• Why Language Models Hallucinate
• Mechanisms of Prompt-Induced Hallucination in Vision-Language Models
• Qwen2.5-Coder Technical Report
• Qwen3 Technical Report
• Prompt Repetition Improves Non-Reasoning LLMs
• PromptArmor: Simple yet Effective Prompt Injection Defenses
• PIShield: Detecting Prompt Injection Attacks via Intrinsic LLM Features

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs