agent/papers/items/2026-2606-20922-think-twice-before-you-act-protecting-llm-agents-against-tool-description-poison.md

Paper: Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated Planning

---

type: paper title: "Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated Planning" authors: Shanghao Shi, Xiao Wang, Chaoyu Zhang, Hao Li, Wenjing Lou, Thomas Hou, Yevgeniy Vorobeychik, Chongjie Zhang, et al. year: 2026 venue: arXiv url: https://arxiv.org/abs/2606.20922 code_url: source: arxiv collected_at: 2026-07-08 published_at: 2026-06-18 updated_at: 2026-06-18 status: queued relevance: high topics:

• agent-evaluation
• agent-safety
• planning
• tool-use methods:

benchmarks:

models:

datasets:

• cs.CR related_concepts:

related_jobs:

related_experiments:

related_projects:

collection_score: 15 collection_queries: planning-agent

One-line Takeaway

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

Why Collected

• matched queries: planning-agent
• inferred topics: agent-evaluation, agent-safety, planning, tool-use
• arXiv categories: cs.CR
• collection score: 15

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/2606.20922