Is this vulnerability remotely exploitable?

It can be, depending on how PPTAgent is integrated. If remote users, external documents, or third-party systems can influence the prompts or context feeding the LLM, they may be able to cause malicious Python code to be generated and evaluated.

What level of access does an attacker need to exploit this?

The attacker needs the ability to shape the LLM's code output, typically by controlling prompts, prompt-injected content (such as slides, notes, or URLs the agent reads), or intermediate tool responses. Direct shell or host access is not required if these channels are reachable.

What is the safest general fix pattern for this issue?

The safest pattern is to remove direct `eval()` of LLM-generated Python entirely and redesign the system around a constrained, declarative interface: the model outputs structured data describing desired actions, and a trusted, audited executor interprets that data using a fixed set of capabilities.

Severity:

HIGH RISK

(8.6)

Code Injection via Unsafe eval (CWE-95) in PPTAgent LLM-Generated PowerPoint Framework

CVE-2026-42079

Published:

May 4, 2026

Status:

Deferred

Summary

Affected Context

Exploit Preconditions

Why This Happens

Potential Impact

General Mitigation Guidance

How Teams Detect Issues Like This

Frequently Asked Questions

Source

NDV

Recent Vulnerabilities

CWE-918

HIGH RISK

HIGH

(8.6)

Server-Side Request Forgery (CWE-918) in OpenClaw Zalo Plugin sendPhoto

Unauthorized Access to Internal Resources via SSRF

May 6, 2026

CWE-863

HIGH RISK

HIGH

(8.8)

Authorization Bypass (CWE-863) in OpenClaw Matrix Room Control Commands

Privilege Escalation via Bypassed Access Control

May 6, 2026

Vulnerability Type

Category:

Code Injection

CWE:

CWE-95

Impact:

Arbitrary Code Execution

Severity:

HIGH RISK

(8.6)

CVSS Version:

3.1

Vulnerability Type

Category:

Code Injection

CWE:

CWE-95

Impact:

Arbitrary Code Execution

Severity:

HIGH RISK

(8.6)

CVSS Version:

3.1

Catch This CVE While Code Is Being Written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Get Started

Ship clean & secure code faster

Start Free Trial

Get Pentest Report

Ship clean & secure code faster

Start Free Trial

Get Pentest Report

Ship clean & secure code faster

Start Free Trial

Get Pentest Report

CodeAnt AI is an offensive and defensive security platform.

Defensive + offensive security.

From first line of code to a pentest
that knows every line of it.

SOC 2 Type II certified | CodeAnt AI is independently audited and compliant with SOC 2 Type II security standards

Ask AI for summary of CodeAnt

Defensive + offensive security.

From first line of code to a pentest
that knows every line of it.

Ask AI for summary of CodeAnt

Defensive + offensive security.

From first line of code to a pentest
that knows every line of it.

Ask AI for summary of CodeAnt

Code Injection via Unsafe eval (CWE-95) in PPTAgent LLM-Generated PowerPoint Framework

Summary

Affected Context

Exploit Preconditions

Why This Happens

Potential Impact

General Mitigation Guidance

How Teams Detect Issues Like This

Frequently Asked Questions

Source

Recent Vulnerabilities

HIGH RISK

HIGH

(8.6)

Server-Side Request Forgery (CWE-918) in OpenClaw Zalo Plugin sendPhoto

HIGH RISK

HIGH

(8.8)

Authorization Bypass (CWE-863) in OpenClaw Matrix Room Control Commands

Vulnerability Type

Vulnerability Type

Vulnerability Type

Catch This CVE While Code Is Being Written

Ship clean & secure code faster

Ship clean & secure code faster

Ship clean & secure code faster

Product

Compare

Company

Developers

Legal

Product

Compare

Company

Developers

Legal

Product

Compare

Company

Developers

Legal