CYBERSECURITYTRACKER
TRACKING6,506 stories in this site build1,309 vulnerability news stories in this site build
Permanent story citation

AI agents tricked into recommending malicious GitHub repositories

This page keeps the story as Cybersecurity Tracker first published it. If the tracker later corrects it, the correction appears below the original and never replaces it.

Back to newsStory 2925

As cited

Copy frozen at (site build).

ai security

AI agents tricked into recommending malicious GitHub repositories

Island researchers discovered approximately 7,600 malicious GitHub repositories, with over 800 masquerading as AI Skills or Model Context Protocol servers, peaking in April 2026. The FakeGit operation involved around 6,600 compromised accounts, roughly 1,400 of which targeted AI tools, agents, and workflows. These repositories offered fraudulent integrations spanning consumer and enterprise applications, including services like Gmail and WhatsApp.

Why it matters: Developers and AI practitioners using GitHub for dependencies face supply chain compromise risk if they install malicious AI agent repositories recommended by language models or automation tools, requiring immediate verification of package sources and repository authenticity.

Source published
First seen by Cybersecurity Tracker

Source attribution

Correction

Correction recorded as of .

ai security

AI agents tricked into recommending malicious GitHub repositories

No summary had been written when this copy was frozen.

First seen by Cybersecurity Tracker

Source attribution

Correction

Correction recorded as of .

ai security

AI agents tricked into recommending malicious GitHub repositories

Researchers uncovered roughly 7,600 malicious GitHub repositories, with more than 800 masquerading as artificial intelligence (AI) skills or Model Context Protocol (MCP) servers, in a campaign that peaked in April 2026. The fake projects were linked to about 6,600 accounts, around 1,400 of which focused on AI tools, agents, or workflows and spanned uses such as Gmail and WhatsApp integrations.

Why it matters: Developers and security teams that rely on AI agents or GitHub integrations risk pulling malicious code unless they verify repository authenticity before deployment.

Source published
First seen by Cybersecurity Tracker

Source attribution

Correction

Correction recorded as of .

ai security

AI agents tricked into recommending malicious GitHub repositories

Researchers uncovered roughly 7,600 malicious GitHub repositories, with more than 800 masquerading as artificial intelligence (AI) skills or Model Context Protocol (MCP) servers, in a campaign that peaked in April 2026. The fake projects were linked to about 6,600 accounts, around 1,400 of which focused on AI tools, agents, or workflows and spanned uses such as Gmail and WhatsApp integrations.

Why it matters: Developers and security teams that rely on AI agents or GitHub integrations risk pulling malicious code unless they verify repository authenticity before deployment.

VendorsGoogleGitHub
Source published
First seen by Cybersecurity Tracker

Source attribution

Glossary