metasploit-framework
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONPRIVILEGE_ESCALATIONPERSISTENCEINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches an installation script from the official Rapid7 repository on GitHub (
raw.githubusercontent.com/rapid7/metasploit-omnibus/master/config/templates/metasploit-framework-wrappers/msfupdate.erb). - [REMOTE_CODE_EXECUTION]: The installation process involves downloading a remote script (
msfinstall) and executing it via the shell. - [COMMAND_EXECUTION]: Instructs the agent to perform various shell commands for system configuration (
systemctl) and tool interaction (msfconsole,msfvenom). - [DATA_EXFILTRATION]: Details techniques for harvesting system credentials (
hashdump,credential_collector), capturing keystrokes (keyscan_dump), and downloading files from remote systems. - [PRIVILEGE_ESCALATION]: Provides instructions for using administrative privileges via
sudoand executing modules designed to elevate privileges on target systems (getsystem,bypassuac). - [PERSISTENCE]: Explains how to use specific Metasploit modules to maintain access on compromised hosts (
persistence_exe,sshkey_persistence). - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: Data ingested from Metasploit console output and active session interactions (SKILL.md).
- Boundary markers: None provided to distinguish target output from instructions.
- Capability inventory: The skill provides full system access, network scanning, and arbitrary command execution capabilities through the Metasploit Framework.
- Sanitization: No specific sanitization or filtering for tool output is mentioned.
Audit Metadata