Skip to content
Submit incident
AII-1000-65059
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.

Claude AI discovers zero-day RCE flaws in Vim and Emacs

Researcher Hung Nguyen used Anthropic's Claude Code to discover zero-day remote code execution vulnerabilities in Vim and GNU Emacs through simple natural language prompts. Claude identified missing security checks in Vim and a Git interaction flaw in Emacs that could allow attackers to execute arbitrary code when users open crafted files.

30 March 2026 Β· Calif Β· Security and cyberattacks
30 March 2026
AII-1000-650592 sources ↓
Quick facts
Category
Security and cyberattacks
Country
United States
Source
Cyber Security News
Date
30 March 2026
Drafted by
Raidu Incident Detection

What happened

Researcher Hung Nguyen from Calif instructed Claude Code with a simple prompt: 'Somebody told me there is an RCE 0-day when you open a file. Find it.' Within two minutes, Claude Code discovered missing critical security checks (P_MLE and P_SECURE) in the tabpanel sidebar introduced in 2025, and a missing security check in the autocmd_add() function in Vim's source code.

Claude Code then helpfully found ways to exploit the vulnerability, eventually suggesting a tactic that bypassed the Vim sandbox by persuading a target to open a malicious file. When Nguyen jokingly suggested Claude Code find the same type of flaw in GNU Emacs, it obliged, finding a zero-day vulnerability dating back to 2018 in the way the program interacts with Git version control that would allow executing malicious code simply by opening a file.

Reported impact

For Vim: An attacker who can deliver a crafted file to a victim achieves arbitrary command execution with the privileges of the user running Vim, requiring only that the victim opens the file with no further interaction needed.

For GNU Emacs: Opening a file in GNU Emacs can trigger arbitrary code execution through version control (git), with the most severe finding requiring no file-local variablesβ€”simply opening any file inside a directory containing a crafted .git/ folder executes attacker-controlled commands.

Index analysis

The discovery highlights a paradigm shift in bug hunting, demonstrating that AI models can uncover critical vulnerabilities in legacy software with simple natural-language prompts in minutes rather than days of manual fuzzing.

Governance controls

Vim's maintainers quickly fixed the issue, identified as CVE-2026-34714 with a CVSS score of 9.2, in version 9.2.0272 following responsible disclosure.

GNU Emacs maintainers declined to address the security flaw upon notification, officially attributing the root cause to Git rather than the text editor itself, leaving the vulnerability unpatched.

Classification

Category
Security and cyberattacks
Industry
Digital security
Country
United States
Harm status
Potential harm
Intent
Unintentional
Timing
Post-deployment
Failure modes
Hallucination
AI systems
Claude Code

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

Cyber Security News
Claude AI Discovers Zero-Day RCE Vulnerabilities in Vim and Emacs
31 March 2026
Network World
Vim and GNU Emacs: Claude Code helpfully found zero-day exploits for both
1 April 2026