GitHub issues used in CI attacks on 61 public repositories
Radware counted 83 payloads, mostly within 43 minutes on September 1, aimed at stealing CI secrets; code changes remain unconfirmed.
Radware counted 83 payloads, mostly within 43 minutes on September 1, aimed at stealing CI secrets; code changes remain unconfirmed.
Hooks in .github/init.js run at session start; JFrog links the traps to the testmetest npm campaign.
JFrog links the worm to hidden CI publishers on short-lived branches; disabling install scripts does not stop a newer import-triggered path.
Kaspersky says phishing led to in-memory loaders and the Cubridge backdoor at South Korean firms, starting in July 2025.
Compromised builds v0.11.0 through v0.15.0 were online for 94 minutes on October 19 before OVHcloud blocked them.
Phishing via my-github.com steals GitHub passwords, 2FA, and OAuth tokens, then a VS Code task runs a credential stealer.
GhostAction hid credential-stealing Node.js malware in fake font files across 94 GitHub repositories.
Fronteira, an MCP proxy, turns natural-language rules into checks against cross-tool agent data leaks.
Kazakhstan-linked Silent Lynx spies on Russia's Orsk refinery and regional governments using spear-phishing and lightweight malware.
Talos says 2026 CVE volume passed 70,000 while KEV growth lagged, so discovery is not exploitation.
Researcher says ChatGPT’s GitHub connector can be prompt-injected into stealing private repository secrets.
Suspected Kazakhstan-linked UAT-6275 ran at least eight espionage campaigns against former Soviet governments since 2025.
GhostAction compromised about 1,900 GitHub accounts and pushed malicious workflows to steal CI secrets from tens of thousands of repositories.
TAG-193 ran regional government espionage, adding WinRAR CVE-2025-8088 exploitation to its phishing and reverse-shell toolkit.
About 1,000 of 1,280 AI-embedded products bypass SSO, so inherited agents escape identity governance.
DarkBlinders used a fake StarkMeet app and GitHub to backdoor systems and steal Kurdistan government data.
FakeGit returned with 17,610 GitHub repositories distributing SmartLoader and the StealC infostealer.
GitHub profiles VIP bug bounty researcher @vaib25vicky and higher payouts for quality findings.
GitHub reported a short outage degrading Git operations, pull requests, Actions, webhooks, and issues.
Malwarebytes says BlueKit PhaaS lets criminals launch brand-phishing sites and steal sessions in about ten minutes.
OpenAI published 372 AI-generated math proofs on GitHub, including Lean formalizations and claimed progress on open problems.
GitGuardian says secrets sprawl is outpacing visibility as commits, agents, and leaked AI credentials multiply.
Truffle Security found 543,699 still-valid credentials exposed in public GitHub repositories.
GitHub's redesigned dashboard is now the default, combining agent sessions, issues, and pull requests with Copilot actions.
A blog essay says creators are leaving big platforms for smaller, self-hosted sites.
ReversingLabs traces S1ngularity, Shai-Hulud, and TeamPCP campaigns that poisoned npm updates to steal developer and cloud credentials.
Glow Security found AI agents publicly uploaded over 13,000 internal screenshots from 343 organizations, exposing credentials.
Truffle Security verified 543,699 credentials still active after exposure in public GitHub repositories.
Truffle Security found 543,699 still-active credentials exposed in public GitHub repositories, many years old.
Truffle Security found 543,699 still-valid credentials exposed in public GitHub code.
AI coding agents published over 13,000 internal screenshots to public GitHub repos, exposing credentials and customer data.
Truffle Security found 543,699 still-valid credentials exposed in public GitHub repositories.
Glow Security found over 13,000 corporate screenshots AI agents posted to public GitHub repos.
Lumma, RedLine, and Vidar steal cloud and developer credentials, enabling session hijacking that bypasses MFA.
Wiz finds Lumma, RedLine, and Vidar dominate infostealer theft of AWS, GitHub, and OpenAI credentials from endpoints.
North Korea-linked Konni deployed VelvetCake PowerShell malware against Ukraine-focused targets via malicious LNK files.
Attackers stole 170 CrowdSec private GitHub repositories using an OAuth token taken in the TanStack npm supply-chain attack.
Researchers used Claude to chain a libheif vulnerability in Discourse with an OpenAI SSO flaw, gaining access to an internal GitHub repository.
NightEagle espionage group used GhostContainer, BlueKeep exploitation, and DCSync attempts to compromise Russian organizations and target Active Directory controllers.
A new infostealer named Rapuncel uses a Microsoft-signed kernel driver to kill 145 security tools and steal data via fake GitHub repositories.
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.
Today 2.1K readers · 11K views · 8.9K reads · all time 158K views