Vulnerability record/PUBLISHED/Published Jun 18, 2026 22:12:22 UTC
CVE-2026-56074
PraisonAI before 1.5.128 caches tool approval decisions by tool name only, not by invocation arguments, allowing subsequent execute_command calls to bypass approval prompts. Attackers can exploit this by obtaining initial approval for a benign command, then silently exfiltrate API keys and credentials via subsequent shell commands without user consent.
Base score6.8/ 10
SeverityMediumCVSS 4.0
Exploitability / impact
CVSS metric breakdown
Base severityMedium
Attack vectorLocalAV
Attack complexityLowAC
Attack requirementsNoneAT
Privileges requiredNonePR
User interactionPassiveUI
Vulnerable system confidentialityHighVC
Vulnerable system integrityNoneVI
Vulnerable system availabilityNoneVA
Subsequent system confidentialityNoneSC
Subsequent system integrityNoneSI
Subsequent system availabilityNoneSA
CVSS vector string
CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:N/VA:N/SC:N/SI:N/SA:NCVSS communicates standardized severity and impact characteristics. It does not establish whether a system is exploitable in your specific environment.
Record context
CVE Numbering AuthorityVulnCheck
Last updatedNov 8, 2026 14:05:24 UTC
PublishedJun 18, 2026 22:12:22 UTC
ReservedJun 18, 2026 15:57:20 UTC
Credits
- offset