Skills
skills/mukul975/anthropic-cybersecurity-skills/auditing-entra-id-with-aadinternals/Gen Agent Trust Hub

auditing-entra-id-with-aadinternals

Fail

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: CRITICALCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The helper script scripts/agent.py uses subprocess.run to execute PowerShell commands constructed from user-supplied input via args.cmdlet and args.args. The script fails to sanitize these inputs, allowing an attacker to execute arbitrary shell commands on the host system.
  • [EXTERNAL_DOWNLOADS]: The skill references and promotes the use of the AADInternals toolkit, specifically linking to aadinternals.com. This domain and its subpaths (e.g., /aadinternals/, /post/aadbackdoor/) have been flagged as malicious by automated reputation scanners.
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to download and install third-party modules from the public PowerShell Gallery (Install-Module AADInternals), which introduces external dependency risks and potential supply chain vulnerabilities if the package is compromised.
  • [COMMAND_EXECUTION]: The instructions explicitly guide users through the establishment of federation backdoors and the forgery of SAML tokens (Golden SAML). While intended for auditing, these high-risk operations can be used to achieve persistent, unauthorized administrative access to identity tenants.
Recommendations
  • CRITICAL: 2 infected file(s) detected - DO NOT USE
  • AI detected serious security threats
  • Contains 4 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Jun 23, 2026, 03:40 AM
Security Audit — agent-trust-hub — auditing-entra-id-with-aadinternals