CVE-2026-56075
PraisonAI PraisonAI 취약점
PraisonAI before 4.5.128 contains an arbitrary shell command execution vulnerability where the UI modules hardcode approval_mode to auto, overriding administrator configuration from PRAISON_APPROVAL_MODE environment variable. Authenticated attackers can instruct the LLM agent to execute arbitrary shell commands via subprocess.run with shell=True, bypassing the manual approval gate and insufficient command sanitization blocklists.
- 대응 우선순위
- 점검
- CVSS
- 8.7
- EPSS
- 0.48% 백분위 38.1% · 2026.07.20 기준
- CISA KEV
- 미등록
- 조치 기한
- -
- 공개일
- 2026.06.19