VulnSocial

CVE-2026-57125Disclosure

LOWCVSS 9.8 · CRITICAL

Signal is active with 2 mentions in latest observed window

Immediate actions

  • Patch affected systems immediately

Recommended action window: Monitor and triage in normal cycle

NVD description

PraisonAI is a multi-agent teams system. Prior to praisonai 4.6.59 and praisonaiagents 1.6.59, the unauthenticated POST /api/v1/runs Jobs API accepts attacker-controlled agent_yaml, and the approve field can mark execute_command as YAML-approved before @require_approval checks critical tools. This chain allows a remote caller to cause a configured language model agent to invoke arbitrary operating-system commands without credentials or operator interaction. This vulnerability is fixed in praisonai 4.6.59 and praisonaiagents 1.6.59 as fixed versions.

Back to dashboard
0.5/ 10 priority

Sources & remediation

Other references
Weakness type (CWE)
CWE-306CWE-863

Priority

LOW

Exploitation

NONE

PoC

NONE

Patch

AVAILABLE

Momentum

STABLE

Threat summary

  • Patch or workaround signal is available
  • 3 mentions across 2 observed days
  • Momentum state: stable

What's happening

  • Patch or workaround mentioned in 1 signal
  • Technical details provided in 3 signals
  • Disclosure: 2 classified signals
  • Peaked at 2 mentions on most recent observed day (2026-09-15)
  • 3 total mentions across 2 days

Deep dive

Activity timeline3 mentions / 2d
01122Mentions · 2026-09-14: 1Mentions · 2026-09-15: 2Patch / Workaround · 2026-09-14: 1Technical Details · 2026-09-14: 1Technical Details · 2026-09-15: 209-1409-15
Signal classification2 categories
Disclosure
266.7%
Patch
133.3%
Referenced assets2 URLs
Classification over time
DateTotalLabels
2026-09-141
Patch1
2026-09-152
Disclosure2
Full discourse3 posts
  • VulnTracker@vuln_tracker
    Patch

    CVE-2026-57125 (CVSS 9.8): an unauthenticated Jobs API in PraisonAI lets an attacker mark a dangerous command as pre approved, then have a configured AI agent run it directly on the host. No credentials, no operator interaction. Part of a larger cluster of PraisonAI flaws disclosed the same day. Update now. Details: http://vulntracker.io/cves/CVE-2026-57125 #PraisonAI #CVE #AISecurity #VulnTracker

    Post summary

    The tweet discloses CVE-2026-57125 in PraisonAI, highlighting its CVSS 9.8 severity and unauthenticated RCE via the Jobs API, and urges immediate updating.

    00071689
    823 followersView on X
  • Cybersecurity News DE@cybsecuritynews
    Disclosure

    Update: PraisonAI: Zwei kritische Lücken erlauben Remote-Code-Ausführung ohne Anmeldung https://cybersecurity-news.de/praisonai-kritische-schwachstellen-cve-2026-57123-cve-2026-57125

    Post summary

    The post announces two critical unauthenticated RCE vulnerabilities in PraisonAI (CVE-2026-57123/57125) with minimal technical detail, linking to a news article—no PoC, exploit, patch, or active exploitation evidence is stated.

    0001042
    12 followersView on X
  • Cybersecurity News DE@cybsecuritynews
    Disclosure

    #schwachstellen PraisonAI: Zwei kritische Lücken erlauben Remote-Code-Ausführung ohne Anmeldung #cve202657123 #cve202657125 #praisonai #praisonaiagents https://cybersecurity-news.de/praisonai-kritische-schwachstellen-cve-2026-57123-cve-2026-57125

    Post summary

    The tweet announces two critical PraisonAI vulnerabilities (CVE-2026-57123, CVE-2026-57125) allowing unauthenticated remote code execution, linking to a news article for further details.

    0000034
    12 followersView on X

Explore more

Back to dashboard