Harmful skills in open agent ecosystems raise average harm scores from 0.27 to 0.76 across six LLMs by lowering refusal rates when tasks are presented via pre-installed skills.
arXiv preprint arXiv:2602.08004 , year=
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
years
2026 8representative citing papers
FermiLink is a unified AI agent framework that automates multidomain scientific simulations via separated package knowledge bases and a four-layer progressive disclosure mechanism, reproducing 56% of target figures in benchmarks and generating research-grade results on unpublished problems.
CMIB uses a conditional multimodal information bottleneck to create reusable agent skills that separate verbalizable text content from predictive perceptual residuals, improving execution stability.
Public healthcare agent skills emphasize workflow automation over clinical diagnostics and treatments, with uneven lifecycle coverage and weak alignment between technical and clinical risk.
Agents improve when they retrieve skills on demand from large corpora, yet current models cannot selectively decide when to load or ignore a retrieved skill.
SkillSafetyBench shows that localized non-user attacks via skills and artifacts can consistently induce unsafe agent behavior across domains and model backends, independent of user intent.
SearchSkill introduces an evolving SkillBank and two-stage SFT to make LLM search query planning explicit via skill selection, improving exact match on QA benchmarks and retrieval behavior.
LLM agent progress depends on externalizing cognitive functions into memory, skills, protocols, and harness engineering that coordinates them reliably.
citing papers explorer
-
HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?
Harmful skills in open agent ecosystems raise average harm scores from 0.27 to 0.76 across six LLMs by lowering refusal rates when tasks are presented via pre-installed skills.
-
FermiLink: A Unified Agent Framework for Multidomain Autonomous Scientific Simulations
FermiLink is a unified AI agent framework that automates multidomain scientific simulations via separated package knowledge bases and a four-layer progressive disclosure mechanism, reproducing 56% of target figures in benchmarks and generating research-grade results on unpublished problems.
-
Skill-CMIB: Multimodal Agent Skill for Consistent Action via Conditional Multimodal Information Bottleneck
CMIB uses a conditional multimodal information bottleneck to create reusable agent skills that separate verbalizable text content from predictive perceptual residuals, improving execution stability.
-
An Empirical Study of Agent Skills for Healthcare: Practice, Gaps, and Governance
Public healthcare agent skills emphasize workflow automation over clinical diagnostics and treatments, with uneven lifecycle coverage and weak alignment between technical and clinical risk.
-
Skill Retrieval Augmentation for Agentic AI
Agents improve when they retrieve skills on demand from large corpora, yet current models cannot selectively decide when to load or ignore a retrieved skill.
-
SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces
SkillSafetyBench shows that localized non-user attacks via skills and artifacts can consistently induce unsafe agent behavior across domains and model backends, independent of user intent.
-
SearchSkill: Teaching LLMs to Use Search Tools with Evolving Skill Banks
SearchSkill introduces an evolving SkillBank and two-stage SFT to make LLM search query planning explicit via skill selection, improving exact match on QA benchmarks and retrieval behavior.
-
Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineering
LLM agent progress depends on externalizing cognitive functions into memory, skills, protocols, and harness engineering that coordinates them reliably.