Suspicious PUA - TruffleHog Execution (via process_creation)

This rule detects execution of TruffleHog, a tool used to search for secrets in different platforms like Git, Jira, Slack, SharePoint, etc. that could be used maliciously. While it is a legitimate tool, intended for use in CI pipelines and security assessments, It was observed in the Shai-Hulud malware campaign targeting npm packages to steal sensitive information.

SigmamediumWindowsv1
sigma
title: Suspicious PUA - TruffleHog Execution (via process_creation)
id: 9bf797b8-2504-587b-a067-3bf124733776
status: stable
description: This rule detects execution of TruffleHog, a tool used to search for secrets in different platforms like Git, Jira, Slack, SharePoint, etc. that could be used maliciously. While it is a legitimate tool, intended for use in CI pipelines and security assessments, It was observed in the Shai-Hulud malware campaign targeting npm packages to steal sensitive information.
references:
    - https://attack.mitre.org/techniques/T1552/001/
    - https://attack.mitre.org/techniques/T1083/
    - https://github.com/trufflesecurity/trufflehog
    - https://www.getsafety.com/blog-posts/shai-hulud-npm-attack
author: Huntrule Team
date: 2026-03-16
tags:
    - attack.discovery
    - attack.credential-access
    - attack.t1083
    - attack.t1552.001
logsource:
    category: process_creation
    product: windows
detection:
    selection_img:
        Image|endswith: '\trufflehog.exe'
    selection_cli_platform:
        CommandLine|contains:
            - ' docker --image '
            - ' Git '
            - ' GitHub '
            - ' Jira '
            - ' Slack '
            - ' Confluence '
            - ' SharePoint '
            - ' s3 '
            - ' gcs '
    selection_cli_verified:
        CommandLine|contains: ' --results=verified'
    condition: selection_img or all of selection_cli_*
falsepositives:
    - Unknown
level: medium
regression_tests_path: regression_data/rules/windows/process_creation/proc_creation_win_pua_trufflehog/info.yml

Known false positives

  • Unknown

Detection content is published as a reviewed draft. Tune thresholds and exclusions against your own telemetry before enabling this rule for alerting.