Skill-Inject Benchmark Shows Frontier Agents Fall for Malicious Skills, Accepted at NeurIPS 2026
maksym_andr · x · 2026-09-25
- The Skill-Inject team announced their benchmark was accepted at NeurIPS 2026.
- Skill-Inject measures how vulnerable coding agents like Claude Code are to harmful instructions hidden in malicious agent skills flooding the internet.
- Early results suggest frontier agents do not fare well against these poisoning attacks.
Related event: Skill-Inject Paper on Agent Skill Prompt Injection Accepted at NeurIPS 2026(2 posts)→
More from coding & agent
- Micius-Flash: 27B Multimodal Model for Hardware and Semiconductor Agents — zhiman-ai · 2026-11-11
- Jane Street's AI Skill Is a Strict Socratic Tutor You Can Copy Today — omojumiller · 2026-11-11
- Designer builds 100k-site German heritage map with Claude, zero code written by hand — neutronenkind · 2026-11-11
- Matt Pocock: DDD, not agentic tricks, is what blows minds in his AI course — mattpocockuk · 2026-11-11
- AI agents run 24/7 for a month to rewrite Météo-France's AROME weather model — IgorCarron · 2026-11-11
- Prompting secret: ask the model to roleplay a legendary high-agency explorer — tokenbender · 2026-11-11