CVE-2026-47391Public exploitNetwork attack vector

Unauthenticated Python execution via A2A eval tool

Published Jul 21, 2026 · Updated Jul 21, 2026

Eval injection in PraisonAI before 4.6.40 allows remote attackers to execute Python through an unauthenticated A2A request. The first-party A2A server example binds to 0.0.0.0 without an auth token, passes message/send input to agent.chat(), and exposes calculate(expression), which sends model-selected input directly to Python eval(). Exploitation requires the official example or a similar public A2A server with an unsafe tool, and code runs with the server process's privileges.

CVSS severity9.8
Critical
EPSS probability1.17%
Next 30 days · Sep 16, 2026
Known exploitationUnconfirmed
Based on sourced intelligence
Hinoki checkNot available
Coverage for this vulnerability

See if you're affected

Explore vulnerability checks for your environment with Hinoki.

Book a demo

Summary

What happened

Eval injection in PraisonAI before 4.6.40 allows remote attackers to execute Python through an unauthenticated A2A request. The first-party A2A server example binds to 0.0.0.0 without an auth token, passes message/send input to agent.chat(), and exposes calculate(expression), which sends model-selected input directly to Python eval(). Exploitation requires the official example or a similar public A2A server with an unsafe tool, and code runs with the server process's privileges.

The record

CVE
CVE-2026-47391
Published
Jul 21, 2026
Updated
Jul 21, 2026
Vendor
PraisonAI
Product
PraisonAI
Classifications
CWE-306, CWE-95, T1190, T1059.006
Attack vector
network
Privileges
unauthenticated

Timeline

How it unfolded

  1. Jul 21, 2026CVE publishedPublication date reported by the CVE source.
  2. Jul 21, 2026Record updatedLatest update available in the CVE record.

Exploitability

Present is not the same as exploitable

Compare your product and version with the public record. A matching version still requires validation against your environment.

Is a vulnerable build present?

Compare these published version ranges with your installed build and any vendor patches.

  1. Affected versionversion=< 4.6.40

What conditions does exploitation require?

Attack vectornetwork
Required privilegesunauthenticated

What is affected?

PraisonAI · PraisonAIversion=< 4.6.40

Published CVSS scores

9.8GitHub, Inc.CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

CVSS describes severity. EPSS estimates exploitation probability.

Attacks

What attackers are doing with it

Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.

Daily unique IPsNo honeypot observations are available for this CVE in the selected window.

No observations available

Sep 10, 2026Sep 16, 2026
Latest reporting daySep 16, 2026
Latest daily unique IPsUnavailable
Prior 30-day averageUnavailable
SourceShadowserver honeypots (KEV)
Vectornetwork
Privilegesunauthenticated
Known exploitationUnconfirmed
Public exploitPublished

Weakness, pattern, technique

CWE-306Missing Authentication for Critical Function
CWE-95Improper Neutralization of Directives in Dynamically Evaluated Code ('Eval Injection')
T1190Exploit Public-Facing Application
T1059.006Command and Scripting Interpreter: Python

Public exploit references

Labels summarize the accepted research assessment. They do not indicate a test against your environment.

Technologies

Your stack

See the directory against your own environment.

Your stack

Check the software in your environment

Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.

Book a demo