CVE-2026-60086: Protection Mechanism Failure in MervinPraison PraisonAI
PraisonAI before 4.6.78 contains a prompt injection defense bypass vulnerability where the injection defense only blocks threats classified as CRITICAL, requiring three or more detector families to match simultaneously. Attackers can craft single or double-vector prompt injections that are classified as HIGH threat level and pass through unblocked to reach the model.
AI Analysis
Technical Summary
CVE-2026-60086 describes a protection mechanism failure in MervinPraison's PraisonAI product prior to version 4.6.78. The prompt injection defense is designed to block threats only if they are classified as CRITICAL and detected by at least three detector families simultaneously. However, prompt injections classified as HIGH threat level can bypass this defense because they do not meet the threshold for blocking. This allows crafted single or double-vector prompt injections to reach the AI model unblocked, potentially undermining the intended security controls.
Potential Impact
Attackers can bypass the prompt injection defense in affected versions of PraisonAI by using injections classified as HIGH threat level, which are not blocked by the current defense logic. This could lead to unauthorized manipulation or influence of the AI model's behavior. The vulnerability has a medium severity rating with a CVSS score of 6.9. There are no known exploits in the wild at this time.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. No official fix or patch information is provided. Users should monitor vendor communications for updates regarding a fix. Until then, be aware of the limitation in the prompt injection defense and consider additional protective measures if feasible.
CVE-2026-60086: Protection Mechanism Failure in MervinPraison PraisonAI
Description
PraisonAI before 4.6.78 contains a prompt injection defense bypass vulnerability where the injection defense only blocks threats classified as CRITICAL, requiring three or more detector families to match simultaneously. Attackers can craft single or double-vector prompt injections that are classified as HIGH threat level and pass through unblocked to reach the model.
CVSS v4.0
Score 6.9medium
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
CVE-2026-60086 describes a protection mechanism failure in MervinPraison's PraisonAI product prior to version 4.6.78. The prompt injection defense is designed to block threats only if they are classified as CRITICAL and detected by at least three detector families simultaneously. However, prompt injections classified as HIGH threat level can bypass this defense because they do not meet the threshold for blocking. This allows crafted single or double-vector prompt injections to reach the AI model unblocked, potentially undermining the intended security controls.
Potential Impact
Attackers can bypass the prompt injection defense in affected versions of PraisonAI by using injections classified as HIGH threat level, which are not blocked by the current defense logic. This could lead to unauthorized manipulation or influence of the AI model's behavior. The vulnerability has a medium severity rating with a CVSS score of 6.9. There are no known exploits in the wild at this time.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. No official fix or patch information is provided. Users should monitor vendor communications for updates regarding a fix. Until then, be aware of the limitation in the prompt injection defense and consider additional protective measures if feasible.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulnCheck
- Date Reserved
- 2026-07-08T12:14:28.344Z
- Cvss Version
- 4.0
- State
- PUBLISHED
- Remediation Level
- null
Threat ID: 6a51062b68715ace43b84829
Added to database: 07/10/2026, 14:48:11 UTC
Last enriched: 07/18/2026, 15:22:13 UTC
Last updated: 08/23/2026, 22:52:12 UTC
Views: 51
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.