ad-trust-attack
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for executing numerous command-line tools used in Active Directory exploitation, such as ldeep, netexec, bloodhound-python, secretsdump.py, and Rubeus.exe across all provided files.
- [PRIVILEGE_ESCALATION]: Detailed procedures are provided for escalating privileges within an AD forest, including the creation of Golden and Diamond tickets to obtain Enterprise Admin rights, and exploiting SID History to bypass security boundaries.
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection: 1. Ingestion points: User-supplied placeholders such as , , , and in SKILL.md and reference files are directly interpolated into shell commands. 2. Boundary markers: None are present to delimit or warn the agent about potentially malicious content in the user-supplied values. 3. Capability inventory: The skill provides access to powerful tools capable of credential dumping, ticket forging, and remote command execution (secretsdump.py, ticketer.py, raiseChild.py, xp_cmdshell). 4. Sanitization: There is no evidence of input validation or escaping for the external data processed by the skill.
Audit Metadata