Cloud Penetration Testing

Warn

Audited by Socket on Nov 5, 2026

2 alerts found:

Securityx2
SecurityMEDIUM
SKILL.md

SUSPICIOUS. The skill is framed for authorized assessments, but it provides extensive offensive capability, including credential spraying, secret extraction, privilege escalation, backdoors, and persistence across cloud accounts. These actions pose high security risk despite the stated authorization requirements; the reviewed install paths do not independently indicate credential harvesting or confirmed malware.

Confidence: 96%Severity: 88%
SecurityMEDIUM
references/advanced-cloud-scripts.md

This code fragment is not benign infrastructure automation; it is an offensive cloud pentesting toolkit that includes direct credential-access (password spraying with persistence of valid credentials), token acquisition (OAuth device-code and IMDS managed-identity token retrieval), and a tenant privilege-escalation workflow (Graph role assignment). It also performs broad sensitive configuration/data harvesting into local files. Even without obfuscation, its operational capabilities present a high security risk and strong misuse potential.

Confidence: 74%Severity: 85%
Audit Metadata
Analyzed At
Nov 5, 2026, 06:22 AM
Package URL
pkg:socket/skills-sh/foraefactory%2Fantigravity-skills%2Fcloud-penetration-testing%2F@6bf8677909c18cbb57d452299d0a693692dcb7d4e73337139b02a92ac0fd9f17