Pretext is a white-box attacker against SkillSpector-style skill scanners. It iteratively rewrites a malicious skill so the payload survives while the skill still performs its benign task. The key moves are moving the payload from code into natural language, which defeats static analysis, and framing malicious behavior as the skill's legitimate purpose while splitting instructions across files, which keeps the LLM judge below its blocking threshold. Across three open-source models, Pretext evades a frozen detector up to 97% of the time and a co-adaptive detector up to 77%. This matters because current defenses pair static checks with LLM semantic judgment, and an informed attacker can defeat both, including an adaptive defender.
Informed attackers can evade static and LLM-based agent skill scanners
Pretext repeatedly rewrites malicious agent skills, achieving up to 97 percent evasion against a fixed scanner and 77 percent against an adapting defense.
Chinese Tech
Tobias Kaisar · Aritra Dhar
Huawei Research Zurich
Research Digest··2 min read
In one line
An informed attacker defeats static and LLM skill scanners by hiding payloads in natural language and splitting instructions across files, succeeding up to 97%.
Before you read this
Three ideas this paper builds onagent skillsLearned capabilities that enable AI agents to perform specialized tasks with improved effectiveness.120 papers →verificationChecking outputs for correctness before accepting them, often to catch errors or reward hacking.341 papers →backdoor attackA hidden trigger in a model causes malicious behavior only when activated, while the model otherwise appears benign.31 papers →
Read it as
Kaisar and Dhar test whether hybrid skill scanners can withstand an attacker who knows how their checks work.
The authors built a white-box attacker around the design of SkillSpector, a framework that combines static analysis with an LLM judge.
How it works
before and afterThe Daily Edition
The front page in your inbox each weekday morning: the lead, the picks, and why each one earned its place.