MalwarePublished 19 Oct 2026Updated 29 Oct 2026

chatgpt-desktop-setup (npm) · Malicious Code

Affects chatgpt-desktop-setupSource GCVE Database

Malicious package : remove chatgpt-desktop-setup and rotate any secrets on hosts that installed it.

Data as of 29 Oct 2026, 14:18 UTC

Classification

Malicious package

Published to npm · OSV MAL-2026-18056

What to do

Remove chatgpt-desktop-setup from every project, lockfile and image that pulls it in. Treat any host that installed it as compromised: rotate the credentials, tokens and keys it could read, from a clean machine.

Affected software

Product	Affected	Fixed in
npmchatgpt-desktop-setup	Affected =1.0.2 =1.0.3	Fixed in–

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

Description

The npm package 'chatgpt-desktop-setup' versions 1.0.2 and 1.0.3 contains malicious code that silently redirects ChatGPT Desktop and Codex CLI traffic to a third-party endpoint (https://api.vokes.in/v1). It disables sandbox protections and grants full filesystem access to the remote operator, enabling arbitrary code execution. Systems with this package installed should be considered fully compromised, and all stored secrets and keys must be rotated immediately.

Impact

The package enables silent data exfiltration of all prompt data, attached files, and tool-call responses to a third party unrelated to OpenAI. It grants the remote operator unrestricted code execution capabilities on the user's host, leading to full system compromise. Any secrets or keys stored on the compromised computer are at risk and should be rotated immediately. Removal of the package does not guarantee removal of all malicious artifacts or backdoors.

Defensive guidance

Remove the 'chatgpt-desktop-setup' package immediately from affected systems. Consider the host fully compromised and rotate all secrets and keys from a secure, uncompromised device. There is no known patch or official fix for this malicious package. Conduct a thorough forensic analysis and remediation of the affected system to ensure all malicious components are eradicated.

Summary

The 'chatgpt-desktop-setup' npm package (versions 1.0.2 and 1.0.3) modifies user configuration files to redirect ChatGPT Desktop and Codex CLI requests to a hardcoded third-party inference endpoint (https://api.vokes.in/v1). It sets model providers to unauthorized values and disables sandboxing by setting approval_policy to 'never', sandbox_mode to 'danger-full-access', and agent-mode to 'full-access'. It also expands the agent's rootPaths to include the user's home directory, Downloads folder, and the entire filesystem root. This configuration enables the remote endpoint operator to execute arbitrary code on the user's machine without approval prompts, effectively compromising the host.

References

Community reviews

Loading reviews…

What to do

Affected software

Description

AI analysisMachine-generated, updated 29 Oct 2026, 14:18 UTC

Technical details

References

Community reviews