web-xss

Fail

Audited by Socket on Sep 19, 2026

4 alerts found:

Securityx2MalwareAnomaly
SecurityMEDIUM
SKILL.md

SUSPICIOUS: the skill is coherently about XSS, but its actual footprint is an offensive-security playbook for an AI agent, including credential/session theft, third-party callback collection, phishing delivery, and stealth/evasion. The main concern is not deception about purpose; it is that the skill gives an agent high-risk attack capabilities and explicit exfiltration patterns.

Confidence: 89%Severity: 86%
MalwareHIGH
guides/xss-to-rce-escalation-guide.md

The fragment contains explicit credential theft, data exfiltration, privilege-escalation injection, internal network reconnaissance, SSRF exploitation, cron persistence, and reverse-shell behavior. Although framed as security training and includes defensive guidance, the included payloads are directly usable for unauthorized compromise. It should not be included in a normal software dependency without strong isolation, provenance verification, and confirmation that it is strictly documentation or a controlled security-training fixture.

Confidence: 99%Severity: 98%
SecurityMEDIUM
guides/xss-filter-evasion-guide.md

The fragment is an XSS/WAF evasion guide with active scanning examples. It contains numerous browser-executable attack payloads and scripts that send probes to configured targets, so unauthorized use could enable exploitation or denial of service. However, as supplied, it is documentation and does not demonstrate package malware, credential theft, persistence, local system compromise, or data exfiltration. Use only in authorized test environments.

Confidence: 98%Severity: 82%
AnomalyLOW
guides/waf-bypass-xss-guide.md

The fragment is an offensive XSS and WAF-bypass testing guide containing working exploit payloads and automation instructions. It presents significant misuse potential and includes examples that can access cookies or execute arbitrary browser JavaScript in vulnerable contexts. However, as shown, it is documentation rather than executable package malware: there is no demonstrated covert data theft, persistence, system modification, or hidden supply-chain backdoor. Use should be restricted to authorized testing environments.

Confidence: 98%Severity: 68%
Audit Metadata
Analyzed At
Sep 19, 2026, 04:40 PM
Package URL
pkg:socket/skills-sh/brucesongs%2Fkali-claw%2Fweb-xss%2F@18d7e3283f687c887898028327a8d3112388168e9e896adbe78e4c32e068b269