Convert indexed Sigma rules into analyst-ready detections.
This studio is built around Phoenix's own rule corpus, not a blank editor. Search by title or rule id, choose a local pySigma backend, then reveal pipelines only when you actually need them.
Indexed Rules
3,779
Ready to search
Backends
3
Loaded from pysigma-node
pySigma Versions
2
Newest: 2.0.1
Translation Workspace
Shape the rule before it leaves Phoenix
Tune Translation
Active Rule
NTLM Hash Leak Via Curl NTLM Authentication
Backends follow their supported pySigma version. Select 1.5.1 for Elasticsearch, Kusto, and other 1.x plugins.
Target Profile
Splunk
Splunk through pysigma-backend-splunk.
Format Mode
Default
Default pySigma output mode for this backend.
NTLM Hash Leak Via Curl NTLM Authentication
Using Splunk · Default · pySigma 2.0.1
Translation controls
Adjust the rule on the left, then regenerate when you want a fresh backend-native query.
title: NTLM Hash Leak Via Curl NTLM Authentication
id: 916eb839-895e-47f8-99ee-3008bf377a3e
status: test
description: |
Detects the use of curl with NTLM authentication and empty credentials (-u :), which can be abused to leak the currently logged-in user's NTLMv2 challenge-response to an
attacker-controlled server, enabling offline cracking or relay attacks.
When no credentials are provided, the Microsoft-shipped curl passes a NULL identity to Windows SSPI, which automatically falls back to the current user's logon session credentials
stored in LSASS — without requiring a plaintext password.
This behavior is exclusive to the curl binary shipped by Microsoft (available since Windows 10 / Windows Server 2019), which is built with SSPI support.
references:
- https://github.com/curl/curl/blob/master/lib/vauth/ntlm_sspi.c#L128-L140
- https://learn.microsoft.com/en-us/windows/win32/secauthn/acquirecredentialshandle--ntlm
- https://curl.se/docs/manpage.html#--ntlm
author: Swachchhanda Shrawan Poudel (Nextron Systems)
date: 2026-06-04
tags:
- attack.credential-access
- attack.t1187
logsource:
category: process_creation
product: windows
detection:
selection_img:
- Image|endswith: '\curl.exe'
- OriginalFileName: 'curl.exe'
selection_ntlm_flag:
CommandLine|contains: '--ntlm'
selection_empty_creds:
CommandLine|re: '(?i)\s(-u|--user)\s*:'
condition: all of selection_*
falsepositives:
- Should be very rare as it's not widely known or used, but could occur in legitimate use cases where curl is used with NTLM authentication and empty credentials, such as in certain scripts or automation tasks.
level: high
regression_tests_path: regression_data/rules/windows/process_creation/proc_creation_win_curl_ntlm_hash_leak_attempt/info.yml
CLI command
Sigma CLI 4.0.0 · Install matching versions, then run from the Phoenix project directory.
python -m pip install 'sigma-cli==4.0.0' 'pysigma==2.0.1' 'pysigma-backend-splunk==3.0.0'
sigma convert --without-pipeline -t splunk -f default 'sigma/rules/windows/process_creation/proc_creation_win_curl_ntlm_hash_leak_attempt.yml'