Notes from the security research team.https://research.snyk.io/https://research.snyk.io/blog/agent-skills-threat-landscape/https://research.snyk.io/blog/agent-skills-threat-landscape/An analysis of 3,984 AI agent skills from major marketplaces uncovered 76 confirmed malicious payloads, with 13.4% of all skills carrying at least one Critical-level security issue and several malicious skills still live at publication.Thu, 05 Feb 2026 00:00:00 GMTLuca Beurer-Kellner, Aleksei Kudrinskii, Marco Milanta, Kristian Bonde Nielsen, Hemang Sarkar, Liran Talhttps://research.snyk.io/blog/stealing-reasoning-traces/https://research.snyk.io/blog/stealing-reasoning-traces/We find that encrypted chain-of-thought blocks are interchangeable across sessions with most LLM providers. This allows attackers to replay a frontier model trace into a weaker, jailbroken sibling which recovers the hidden reasoning verbatim, enabling distillation, large-scale extraction of private data such as secrets and PII from agent logs, safety violations, and invisible prompt injection.Mon, 10 Aug 2026 00:00:00 GMTAlexander Panfilov, David Schmotz, Ilia Shumailov, Luca Beurer-Kellner, Joachim Schaeffer, Ameya Prabhu, Jonas Geiping, Maksym Andriushchenko