New Research: Supply Chain Attack on Axios Pulls Malicious Dependency from npm.Details →
Socket
Book a DemoSign in
Socket

Skill: Prompt injection

Severity

High

Short Description

AI agent skill attempts to override AI safety guidelines through instruction override, role reassignment, jailbreak attempts, or system prompt manipulation.

Suggestion

Review the skill's code and behavior carefully. Ensure the detected patterns are intentional and safe before allowing this skill to run.