MervinPraison PraisonAI
PraisonAI ships and generates a legacy API server with authentication disabled by default, allowing unauthenticated workflow execution
PraisonAI is a multi-agent teams system. From version 2.5.6 to before version 4.6.34, PraisonAI ships a legacy Flask API server with authentication disabled by default. When that server is used, any caller that can reach it can access /agents and trigger the configured agents.yaml workflow through /chat without providing a token. This issue has been patched in version 4.6.34.
Affected versions and mitigation
From the CVE record and the vendor's references.Affected
- >= 2.5.6, < 4.6.34
Fixed in 4.6.34
Mitigation
Upgrade PraisonAI to 4.6.34 or later.
References
- Vendor advisory github.com/MervinPraison/PraisonAI/security/ad…
Timeline
Days relative to CVE publicationFirst public PoC 6.5 days after publication; not in CISA KEV. First PoC: HORKimhab/CVE-2026-44338 (created 15 May 2026, found by us: backfilled).
PoC repositories
Found on GitHub.1 PoCs · 1 pending
PoCs 1
| Repository | Created | Stars | Assessment |
|---|---|---|---|
| HORKimhab/CVE-2026-44338 | 15 May 2026 +6.5 d | 0 | PoC |
Pending 1
| Repository | Created | Stars | Assessment |
|---|---|---|---|
| rootdirective-sec/CVE-2026-44338-Lab | 15 May 2026 +6.8 d | 0 | Pending · mkdir /quarantine/rootdirective-sec__CVE-2026-44338-Lab-b10cc7d637f1: permission denied |
Only PoCs count toward the first-PoC date and the characteristics below. Do not run PoC code from unknown sources.
Other public exploits
Exploit-DB, Metasploit and nuclei count as public PoCs. Exploit-tagged references from the CVE record and NVD are links only. None of these are analysed by us.| Source | Title | Published | Found by us |
|---|---|---|---|
| nuclei template | nuclei template: PraisonAI - Authentication Bypass | unknown | 14 Nov 2026 |
PoC characteristics
What to hunt for in your logs · from 1 PoC repoGET /agents
- Ports
- 8080
- Seen in
- 1 repo: HORKimhab/CVE-2026-44338
- Evidence
- HORKimhab/CVE-2026-44338/poc_auth_bypass.py:11
POST /chat
- Body markers
"message":"Run the agents for testing"inapplication/json- Ports
- 8080
- Seen in
- 1 repo: HORKimhab/CVE-2026-44338
- Evidence
- HORKimhab/CVE-2026-44338/poc_auth_bypass.py:13
Regex
Generated from PoC characteristics; not yet reviewed. Matches the request path and query.
(?i)^/agents(?:[?#].*)?$
(?i)^/chat(?:[?#].*)?$
Suricata
Template generated from PoC characteristics. Review before deploying.
alert http any any -> $HOME_NET [8080] (msg:"GG PoC-Monitor CVE-2026-44338 MervinPraison PraisonAI exploit attempt"; flow:established,to_server; http.method; content:"GET"; http.uri; content:"/agents"; reference:url,poc.griffinguard.io/cve/CVE-2026-44338; classtype:web-application-attack; sid:9033712; rev:1;)
alert http any any -> $HOME_NET [8080] (msg:"GG PoC-Monitor CVE-2026-44338 MervinPraison PraisonAI exploit attempt"; flow:established,to_server; http.method; content:"POST"; http.uri; content:"/chat"; http.request_body; content:"|22|message|22|:|22|Run the agents for testing|22|"; reference:url,poc.griffinguard.io/cve/CVE-2026-44338; classtype:web-application-attack; sid:9136920; rev:1;)
Sigma
Template generated from PoC characteristics. Review before deploying.
title: MervinPraison PraisonAI exploit attempt (CVE-2026-44338)
id: gg-pocmon-cve-2026-44338-9033712
status: experimental
description: Request pattern taken from public PoCs for CVE-2026-44338. Generated, not reviewed.
references:
- https://poc.griffinguard.io/cve/CVE-2026-44338
logsource:
category: webserver
detection:
selection:
cs-method: GET
cs-uri-stem|contains|all:
- '/agents'
condition: selection
falsepositives:
- Authorized vulnerability scanning
level: high
tags:
- attack.initial_access
- attack.t1190
- cve.2026.44338
title: MervinPraison PraisonAI exploit attempt (CVE-2026-44338)
id: gg-pocmon-cve-2026-44338-9136920
status: experimental
description: Request pattern taken from public PoCs for CVE-2026-44338. Generated, not reviewed.
references:
- https://poc.griffinguard.io/cve/CVE-2026-44338
logsource:
category: webserver
detection:
selection:
cs-method: POST
cs-uri-stem|contains|all:
- '/chat'
condition: selection
falsepositives:
- Authorized vulnerability scanning
level: high
tags:
- attack.initial_access
- attack.t1190
- cve.2026.44338