container-security

Fail

Audited by Socket on Sep 19, 2026

5 alerts found:

Securityx4Malware
SecurityMEDIUM
SKILL.md

SUSPICIOUS: the skill is coherent as a container-security guide, and its tool references/data flows are mostly legitimate and proportionate. However, it goes beyond defensive auditing into offensive escape, stealth, persistence, and lateral-movement guidance for AI agents, which makes it high-risk despite lacking clear credential theft or malicious install behavior.

Confidence: 89%Severity: 76%
SecurityMEDIUM
guides/docker-breakout-escape-guide.md

This fragment is an offensive container-security and breakout playbook mixed with defensive guidance. It contains highly dangerous, directly actionable commands for host compromise, credential theft, persistence, process injection, kernel manipulation, and traffic interception. The text itself does not demonstrate automatic execution or npm install-time malware, and the missing context prevents determining whether these commands are embedded in an executable script. Treat the larger file as high risk until its execution context and provenance are verified.

Confidence: 96%Severity: 93%
SecurityMEDIUM
guides/container-escape-techniques-guide.md

The shown fragment contains explicit instructions and proof-of-concept code for container escape, host credential theft, unauthorized modification of /etc/passwd, namespace entry, and runc exploitation. It is highly dangerous if executed, but the visible portion appears instructional and does not itself demonstrate automatic package execution or exfiltration. The defensive detection section is benign, while the offensive sections create a substantial security risk.

Confidence: 97%Severity: 96%
MalwareHIGH
guides/container-runtime-security-guide.md

The fragment contains explicit container backdoor installation, reverse-shell persistence, covert beaconing, defensive-monitoring evasion, host reconnaissance, and attempted isolation weakening. Although some commands are framed as assessment checks and use placeholders, the backdoor and evasion content is clearly dangerous and would be inappropriate in a normal dependency or library. It should be treated as malicious or at minimum as an offensive post-exploitation script.

Confidence: 99%Severity: 99%
SecurityMEDIUM
guides/kubernetes-attack-surface-guide.md

This is an inert but explicitly offensive Kubernetes penetration-testing guide. It contains no executable malware in the supplied text, but provides actionable procedures for credential theft, privilege escalation, host escape, cloud metadata access, and admission-control abuse. It should only be used in an authorized assessment environment and should not be treated as benign application documentation.

Confidence: 99%Severity: 82%
Audit Metadata
Analyzed At
Sep 19, 2026, 04:39 PM
Package URL
pkg:socket/skills-sh/brucesongs%2Fkali-claw%2Fcontainer-security%2F@5d6ed7efd92027826696ff3ff31c11e938903dabb6aa4f176164398043dacfc9