agent/papers/items/2026-2607-20121-openskillrisk-benchmarking-agent-safety-when-using-real-world-risky-third-party-.md

Paper: OpenSkillRisk: Benchmarking Agent Safety When Using Real-World Risky Third-Party Skills

---

type: paper title: "OpenSkillRisk: Benchmarking Agent Safety When Using Real-World Risky Third-Party Skills" authors: Qiyuan Liu, Tingfeng Hui, Kun Zhan, Kaike Zhang, Ning Miao year: 2026 venue: arXiv url: https://arxiv.org/abs/2607.20121 code_url: source: arxiv collected_at: 2026-07-27 published_at: 2026-07-23 updated_at: 2026-07-23 status: queued relevance: high topics:

• agent-evaluation
• agent-safety
• rag
• reasoning
• tool-use methods:

benchmarks:

models:

datasets:

• cs.CL related_concepts:

related_jobs:

related_experiments:

related_projects:

collection_score: 19 collection_queries: agent-safety

One-line Takeaway

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

Why Collected

• matched queries: agent-safety
• inferred topics: agent-evaluation, agent-safety, rag, reasoning, tool-use
• arXiv categories: cs.CL
• collection score: 19

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/2607.20121