Duplicate Advisory: PraisonAI: Prompt-injection defense blocks only when 3+ detector families fire simultaneously; realistic single-vector injections pass through unblocked
Description
PraisonAI versions up to 4.6.77 have a prompt injection defense vulnerability where the defense only blocks threats classified as CRITICAL, requiring three or more detector families to trigger simultaneously. This allows attackers to bypass the defense with single or double-vector prompt injections classified as HIGH threat level, which pass through unblocked.
CVSS v3.1
Score 5.3medium
Affected software
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
Weaknesses
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
PraisonAI before version 4.6.78 contains a vulnerability in its prompt injection defense mechanism. The defense only blocks injection attempts when three or more detector families simultaneously classify the threat as CRITICAL. However, attackers can craft prompt injections that trigger only one or two detector families, resulting in a HIGH threat classification that is not blocked. This bypass allows such injections to reach the model unimpeded.
Potential Impact
The vulnerability allows prompt injection attacks classified as HIGH threat level to bypass the defense mechanism, potentially enabling attackers to influence or manipulate the AI model's behavior. There is no indication of confidentiality or availability impact. No known exploits in the wild have been reported.
Mitigation Recommendations
A fix is available in PraisonAI version 4.6.78 and later. Users should upgrade to version 4.6.78 or newer to remediate this vulnerability.
Technical Details
- Gcve Source
- db.gcve.eu
- Osv Id
- GHSA-5c3v-h6hw-4gx7
- Osv Schema Version
- 1.4.0
- Ecosystems
- ["PyPI"]
- Database Specific Severity
- MODERATE
- Cvss Version
- 3.1
Threat ID: 6ac80f0e2cdf04f65639049f
Added to database: 10/08/2026, 21:45:50 UTC
Last enriched: 10/08/2026, 21:50:22 UTC
Last updated: 10/08/2026, 21:50:22 UTC
Views: 2
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.
External Links
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.