agent/papers/items/2026-2607-02857-mosaic-knowledge-guided-cli-command-composition-attack-in-llm-coding-agents.md

Paper: MOSAIC: Knowledge-Guided CLI Command Composition Attack in LLM Coding Agents

---

type: paper title: "MOSAIC: Knowledge-Guided CLI Command Composition Attack in LLM Coding Agents" authors: Jiangrong Wu, Huaijin Wang, Yihao Zhang, Yuhong Nan, Shuai Wang year: 2026 venue: arXiv url: https://arxiv.org/abs/2607.02857 code_url: source: arxiv collected_at: 2026-07-08 published_at: 2026-07-03 updated_at: 2026-07-03 status: queued relevance: high topics:

• agent-evaluation
• agent-safety
• coding-agent
• computer-use
• workflow-agent methods:

benchmarks:

models:

datasets:

• cs.CR
• cs.SE related_concepts:

related_jobs:

related_experiments:

related_projects:

collection_score: 17 collection_queries: agent-evaluation, coding-agent

One-line Takeaway

Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.

Why Collected

• matched queries: agent-evaluation, coding-agent
• inferred topics: agent-evaluation, agent-safety, coding-agent, computer-use, workflow-agent
• arXiv categories: cs.CR, cs.SE
• collection score: 17

Review Checklist

• Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
• Does it include a benchmark, dataset, code, or reproducible experimental setup?
• Should it be promoted from queued to skimmed or summarized?

Links

• arXiv: https://arxiv.org/abs/2607.02857