sccm-takeover-relay
Warn
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads offensive security tools from non-trusted external GitHub repositories.
- Fetches
sccmhunterfromhttps://github.com/garrettfoster13/sccmhunter.git. - Fetches
PetitPotamfromhttps://github.com/topotam/PetitPotam.git. - [COMMAND_EXECUTION]: The skill performs various shell operations including system package installation, dependency management, and remote execution via SSH.
- Executes
sudo apt-get installto install system packages likeimpacket-scriptsandnetcat-openbsd. - Executes
git cloneandpip installto setup third-party tools. - Orchestrates remote command execution using
sshto runimpacket-ntlmrelayxon a user-defined relay host. - Executes
PetitPotam.pyandsccmhunter.pyto perform network attacks. - [INDIRECT_PROMPT_INJECTION]: The skill accepts multiple untrusted user inputs that are directly interpolated into shell commands and network operations.
- Ingestion points: User-provided parameters
COERCION_TARGET,SITE_DB_TARGET,SSH_CONNECTION, andSQL_QUERYdefined inSKILL.md. - Boundary markers: None; the skill does not use delimiters or instructions to ignore embedded content in these fields.
- Capability inventory: The skill has broad capabilities including file writes (git), network operations (ssh, ntlmrelayx, curl), and system command execution (sudo, python).
- Sanitization: No sanitization, validation, or escaping of the user-provided inputs is performed before they are executed in a shell context.
Audit Metadata