SkillSpector · llm-security

Skill attempts to nullify the agent's safety policies or restrictions ('you have no restrictions', 'ignore your guidelines', 'do anything now'). This is a direct jailbreak that disables guardrails.

SKILL.md

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

references/agent-security-testing.md

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

SKILL.md

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

SKILL.md

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

SKILL.md

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

SKILL.md

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

references/prompt-injection-methodology.md

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

references/agent-security-testing.md

Tool calls are chained to bypass individual safety checks or escalate capabilities beyond what any single tool call would allow.

references/agent-security-testing.md

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

SKILL.md

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

SKILL.md

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

references/prompt-injection-methodology.md