AgentBaiting Campaign Exploits AI Skills to Distribute SmartLoader Malware

www.island.io

AgentBaiting Campaign Exploits AI Skills to Distribute SmartLoader Malware

First seen 21 Jul 2026, 19:22 UTC • •64.5

Article Content

1 / 4

Browse articles

ThreatCluster •July 22, 2026 at 18:56 UTC

•AgentBaiting uses 800 fake AI Skills to deliver SmartLoader malware.
•The operation exploits trusted GitHub projects and public AI registries.
•7,600 repositories are implicated in the malware distribution campaign.

The AgentBaiting campaign has emerged as a significant threat, utilizing 800 fake AI Skills and Model Context Protocol (MCP) servers to deliver SmartLoader malware. This operation leverages trusted GitHub projects and public capability catalogs to disguise malicious activities. The malware delivery method exploits the growing trust in AI integrations, turning them into vectors for cyberattacks. The campaign is notable for its scale, with 7,600 repositories involved in the operation. Affected systems include those relying on AI capabilities and MCP workflows. The current status of the campaign indicates ongoing exploitation, with no immediate resolution reported. Security professionals are advised to remain vigilant against this evolving threat.

Start a free Starter trial for enhanced analysis

Ask AI about this cluster

Answers cite the sources they use

Updated 96d ago How this analysis works

Timeline

2026-07-21

AgentBaiting campaign identified

Malware operators are using fake AI Skills and MCP servers to distribute SmartLoader malware through trusted platforms.

2026-07-21

Malware delivery method detailed

The campaign turns trusted AI integrations into vectors for cyberattacks, affecting numerous systems relying on AI capabilities.

Cybersecuritynews

More articles in this cluster (3)

Threat Graph

8 entities, 6 inferred relationships (STIX 2.1).

The full threat graph for this incident is free in the 7-day Starter trial.

Create free account

No card · 30 seconds

See everything included

Technique (3)

Malware
Supply Chain Attack
T1195 - Supply Chain Compromise

Campaign (2)

AgentBaiting Campaign
SmartLoader

Malware (1)

StealC

Affected Platform (1)

GitHub

Tool (1)

FakeGit

Relationships

AgentBaiting Campaign uses StealC
SmartLoader uses StealC
Malware targets GitHub
Supply Chain Attack targets GitHub
T1195 - Supply Chain Compromise targets GitHub
AgentBaiting Campaign related to FakeGit

Following this threat?

Track StealC in your own feed — you're alerted when they show up in new reporting, leak sites or exploitation.

Free account · no card needed

Continue Reading

AI-Assisted Phishing Targets Taiwan Research Institutions An advanced persistent threat (APT) spear-phishing campaign has been identified targeting individuals affiliated with Taiwan research organizations. The campaign utilized AI-assisted content generation to create highly personalized phishing emails that impersonated reputable academic and policy institutions. Attack…

FakeGit Malware Campaign Resurfaces with 17,610 Malicious Repositories The FakeGit malware campaign has reactivated, distributing SmartLoader malware via over 17,610 fake repositories on GitHub. This resurgence, noted by researchers from Apiiro, began on October 4, 2026, and has already seen the creation of more than 13,000 repositories in just 34 hours. The malicious repositories employ…