Skip to main content
VulnerabilityPublished Updated

CVE-2026-34935: PraisonAI · OS Command Injection in MCPHandler.parse_mcp_command()

Affects praisonaiSource GCVE Database

0
Compare
Authoritative record:CVE-2026-34935· Critical 9.8

Not known to be exploited · CVSS 9.8 Critical : upgrade to 4.5.69 as a priority.

Data as of 16 Nov 2026, 16:37 UTC

CVSS 3.1 base score
9.8Critical
How severe the flaw is on the 0–10 CVSS scale.
Patch
Available – fixed in 4.5.69
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

Exploitation evidence

  1. Disclosed: evidence found.

    CVE published 3 Apr 2026

  2. Detectable: no evidence on record.

    No public nuclei template indexed

  3. Weaponized: no evidence on record.

    No public exploit code indexed in Metasploit or Exploit-DB

  4. Exploited in the wild: no evidence on record.

    Not in CISA KEV and no source reports exploitation

  5. Observed by OffSeq Mirage: no evidence on record.

    No attack traffic recorded by OffSeq Mirage sensors

A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), the Metasploit, Exploit-DB and nuclei-templates index as of 15 Nov 2026.

What to do

Upgrade praisonai to 4.5.69.

Affected software

ProductAffectedFixed in
PyPIpraisonaiAffected
  • >=4.5.15 <4.5.69
Fixed in
  • 4.5.69

Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.

AI-suggested

Upgrade praisonai to version 4.5.69 or later, where the vulnerability is fixed by enforcing a command allowlist for the --mcp argument. Until upgraded, do not allow untrusted input to control the --mcp CLI argument. No other mitigations are documented or recommended.

Description

PraisonAI versions from 4.5.15 up to but not including 4.5.69 contain an OS command injection vulnerability in the MCPHandler.parse_mcp_command() function. The vulnerability arises because the --mcp CLI argument is passed without validation or sanitization to a shell command execution function, allowing arbitrary OS commands to be executed as the process user. This issue is fixed in version 4.5.69 by introducing a command allowlist that restricts executable commands. Exploitation requires no authentication and can lead to full system compromise under the process user context.

References

Community reviews

Loading reviews…