Shell Spawned by Celery Worker Process
experimentalmediumLog-Quelle: linux · process_creationAutor Jennifer BatinaDatum 2026-10-17
01
Beschreibung
Was dies ist, in Kürze.
Detects a system shell interpreter being spawned directly by a Celery worker process. This behavior indicates a Python background task execution framework is evaluating raw string input through a shell execution context, which is highly indicative of successful command or prompt injection exploitation against an LLM agent task.
02
Tags
attack.executionattack.t1059.004
03
Regelquelle (YAML)
title: Shell Spawned by Celery Worker Process
id: 904f6442-3b57-4b3f-987c-1559f58ce8d8
status: experimental
description: |
Detects a system shell interpreter being spawned directly by a Celery worker process.
This behavior indicates a Python background task execution framework is evaluating raw string input through a shell execution context,
which is highly indicative of successful command or prompt injection exploitation against an LLM agent task.
references:
- https://docs.celeryq.dev/en/stable/userguide/tasks.html
- https://genai.owasp.org/llmrisk/llm01-prompt-injection/
- https://labs.cloudsecurityalliance.org/research/csa-research-note-promptware-c2-agent-exploitation-20260406/
- https://nvd.nist.gov/vuln/detail/CVE-2021-23727
- https://owasp.org/www-project-top-10-for-large-language-model-applications/
- https://security.snyk.io/vuln/SNYK-PYTHON-CELERY-2314953
- https://snyk.io/blog/python-rce-vulnerability/
- https://unit42.paloaltonetworks.com/ai-agent-prompt-injection/
- https://www.microsoft.com/en-us/security/blog/2026/05/07/prompts-become-shells-rce-vulnerabilities-ai-agent-frameworks
author: Jennifer Batina
date: 2026-10-17
tags:
- attack.execution
- attack.t1059.004
logsource:
category: process_creation
product: linux
detection:
selection_img:
Image|endswith:
- '/ash'
- '/bash'
- '/busybox'
- '/csh'
- '/dash'
- '/fish'
- '/ksh'
- '/sh'
- '/tcsh'
- '/zsh'
ParentCommandLine|contains: 'celery'
selection_parent:
- ParentImage|endswith:
- '/python'
- '/python3'
- ParentImage|contains: '/python3.'
condition: all of selection_*
falsepositives:
- Legacy or custom administrative tasks deliberately hardcoded with shell=True or os.system execution inside Celery task functions.
- Media processing, deployment, or reporting pipelines that intentionally shell out to external tools (e.g. ffmpeg, git) from within a task.
level: medium
04
erkennt Techniken (1)
ATT&CK-Techniken, die dies ermöglicht oder abbildet.
05
Referenzen
Externe Belege — Patches, Meldungen, Analysen.
- https://docs.celeryq.dev/en/stable/userguide/tasks.html
- https://genai.owasp.org/llmrisk/llm01-prompt-injection/
- https://labs.cloudsecurityalliance.org/research/csa-research-note-promptware-c2-agent-exploitation-20260406/
- https://nvd.nist.gov/vuln/detail/CVE-2021-23727
- https://owasp.org/www-project-top-10-for-large-language-model-applications/
- https://security.snyk.io/vuln/SNYK-PYTHON-CELERY-2314953
- https://snyk.io/blog/python-rce-vulnerability/
- https://unit42.paloaltonetworks.com/ai-agent-prompt-injection/