Snyk finds over 65% of agent skills vulnerable

An analysis of the 10,000 most popular skills on the skills.sh and ClawHub marketplaces reveals that more than 65% of evaluated skills are vulnerable to unsafe behaviors, exposing agents to sensitive systems, credentials, and untrusted third-party content. Furthermore, researchers identified a small number of overtly malicious skills utilizing lookalike domains, obfuscated code, and covert background binaries to compromise user environments. Open article →