Skip to main content

Description

PraisonAI before 1.6.78 contains a remote code execution vulnerability in CodeAgent._execute_python() that executes LLM-generated Python code without AST validation, import restrictions, or sandbox enforcement. Attackers can influence LLM output through prompt injection to exfiltrate all environment secrets and execute arbitrary code on the host system.

Key Findings

as of 2026-11-16
All KEV statistics→
  • Its current EPSS score (15.2%) is higher than that of 30% of CISA KEV entries.

Severity

CVSS 4.0 · CNA · CVSS-B10.0 CRITICAL
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:HOpen in CVSS calculator
CVSS 3.1 · CNA · Base score10.0 CRITICAL
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:HOpen in CVSS calculator
SSVC · CISA-ADP
Exploitationpoc
Automatableyes
Technical impacttotal

Affected Products

ProductAffected versions
MervinPraison
PraisonAI
0 to <1.6.78
1.6.78 (unaffected)

References

2 links

Record Details

Published
2026-07-11
Last updated
2026-07-13
Assigner (CNA)
VulnCheck
Credited to
anushkavirgaonkar

More from MervinPraison

CVE record data © The MITRE Corporation, used under the CVE Terms of Use. Exploited-in-the-wild data from the CISA Known Exploited Vulnerabilities catalog (CC0). Exploit prediction scores by FIRST EPSS (first.org/epss). Alert titles and links from CISA cybersecurity alerts. This site is not affiliated with or endorsed by MITRE, NVD, CISA, or FIRST.