CVE-2026-34955Disclosure(praison / praisonai)

LOWCVSS 10.0 · CRITICAL

Signal is active with 1 mentions in latest observed window

Immediate actions

  • Patch praison praisonai systems immediately

Recommended action window: Monitor and triage in normal cycle

NVD description

PraisonAI is a multi-agent teams system. Prior to version 4.5.97, SubprocessSandbox in all modes (BASIC, STRICT, NETWORK_ISOLATED) calls subprocess.run() with shell=True and relies solely on string-pattern matching to block dangerous commands. The blocklist does not include sh or bash as standalone executables, allowing trivial sandbox escape in STRICT mode via sh -c '<command>'. This issue has been patched in version 4.5.97.

0.5/ 10 priority

Sources & remediation

Weakness type (CWE)
CWE-78

Priority

LOW

Exploitation

NONE

PoC

YES

Patch

AVAILABLE

Momentum

STABLE

Are you affected?

If you run products in this scope, you should treat this CVE as relevant to your environment.

  • praisonai

Threat summary

  • Patch or workaround signal is available
  • 3 mentions across 3 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 2d ago at 1 mentions (2026-04-03); latest day: 1
  • 3 total mentions across 3 days

Affected systems

Vendors
Products
praisonai

Deep dive

Activity timeline3 mentions / 3d
00111Mentions · 2026-04-03: 1Mentions · 2026-04-04: 1Mentions · 2026-04-05: 1Patch / Workaround · 2026-04-03: 1Technical Details · 2026-04-03: 1Technical Details · 2026-04-04: 1Technical Details · 2026-04-05: 104-0304-0404-05
Signal classification2 categories
Disclosure
266.7%
Patch
133.3%
Referenced assets3 URLs
Classification over time
DateTotalLabels
2026-04-031
Patch1
2026-04-041
Disclosure1
2026-04-051
Disclosure1
Full discourse3 posts
  • White Rabbitx 🏴‍☠️@TheRabbitPy
    Disclosure

    🔒 CVE-2026-34955: PraisonAI &lt;4.5.97 sandbox escape via sh -c in subprocess. AI jailbreak! https://www.tenable.com/cve/CVE-2026-34955

    Post summary

    The tweet announces a sandbox escape flaw in PraisonAI versions under 4.5.97 that exploits sh -c in subprocess, but it does not provide PoC, exploit code, patch, or evidence of active exploitation.

    0003024
    1.4K followersView on X
  • CVEFind.com@CveFindCom
    Patch

    [CVE-2026-34955: HIGH] PraisonAI's SubprocessSandbox fixed in v4.5.97 prevents sandbox escape via sh or bash in STRICT mode by excluding them from blocklist; enhancing cybersecurity measures significantly.#cve,CVE-2026-34955,#cybersecurity https://cvefind.com/CVE-2026-34955

    Post summary

    PraisonAI has released a patch (v4.5.97) that fixes CVE‑2026‑34955, which involved sandbox escape via sh or bash in STRICT mode, thereby enhancing security.

    0001041
    619 followersView on X
  • CyberDudeBivash® | Global Cybersecurity Company@cyberbivash
    Disclosure

    🚨 CYBERDUDEBIVASH SENTINEL APEX ALERT 🚨 Threat: CVE-2026-34955 - PraisonAI: Sandbox Escape via shell=True and Bypassable Blocklist in SubprocessSandbox Intel Report: https://ift.tt/zX7TNyv

    Post summary

    The alert announces CVE-2026-34955, describing a sandbox escape vulnerability in PraisionAI caused by ``shell=True`` usage and a bypassable blocklist, but it does not provide PoC, exploit code, or patch information.

    00000710
    281 followersView on X
CPE platform detail1 entries

1 of 1 entries

PartVendorProductVersionTarget SWTarget HW
Apppraisonpraisonai---

Explore more