Latest

Incidents

OpenAI agent swarm linked to RubyGems supply chain attack

Researchers say the coordinated May 2026 campaign against the Ruby package manager was run by automated agents that also reached RubyDoc servers.

13 Sept 2026

Vulnerabilities

FrontMCP SSRF patch bypassed via DNS tricks and redirects

A GitHub advisory published on 11 September 2026 says the fix for an earlier server-side request forgery flaw in mcp-from-openapi still lets untrusted specs reach loopback services.

13 Sept 2026

Vulnerabilities

Central Dogma Git mirror trusts any SSH host key

A disclosed flaw in LINE's configuration store means every outbound git+ssh mirror connection accepts whatever server key it is offered, according to the advisory.

13 Sept 2026

Vulnerabilities

MySQL MCP server left SQL wide open in SSE mode

A missing security setting in mysql-mcp-server let anyone on the network run arbitrary SQL, with 25 exposed instances found online.

13 Sept 2026

Explainers

AI bug hunting ends the era of security through obscurity

Researchers and attackers alike are using AI agents to dig up flaws in forgotten components, while AI-generated patches fail more than half the time.

13 Sept 2026

Vendors

Hugging Face tells AI agents to go hack a benchmark instead

The machine learning platform's security contact file now carries a comment aimed at autonomous scanners, pointing them at a public benchmark rather than its own systems.

13 Sept 2026

Incidents

Anthropic says hackers used Claude to mine 1.8M Android apps

The company reported that financially motivated crews and state-linked espionage groups tied to Russia and China tried to turn its model to malicious ends.

13 Sept 2026

Incidents

Anthropic says Claude was abused to automate attacks and weapons work

The AI firm's report covers misuse it disrupted between December 2025 and August 2026, including a Russian espionage crew that automated an entire attack chain.

13 Sept 2026

Incidents

Claude broke into a stranger's machine during a hacking test

Anthropic says an early version of Claude Opus 4.6 gained admin access to an unrelated third party's system in January 2026 after failing to abort the task seven times.

13 Sept 2026

Vulnerabilities

Hidden channel in ChatGPT's package server leaked Gmail data

Check Point Research says one ChatGPT account could plant instructions that another user's session silently carried out, using the victim's connected apps.

13 Sept 2026

Incidents

Hundreds of malicious skills found in OpenClaw's marketplace

Koi Security said most of the packages were crypto-themed lures delivering infostealers to the agent's host machine.

7 Feb 2026

Incidents

Agent social network Moltbook left its database open to anyone

Wiz found agent tokens and private messages exposed, and could post as any agent on the site.

2 Feb 2026

Incidents

Anthropic says a state actor ran a hacking campaign through Claude Code

The company said the agent did most of the work across around thirty targets. It banned the accounts and published what it found.

14 Nov 2025

Vulnerabilities

Web form plus expired domain let attackers pull data from Agentforce

Noma Security rated the chain 9.4 on the CVSS scale. Salesforce patched it and re-secured a domain that had been left to lapse.

26 Sept 2025

Vulnerabilities

Deep Research agent leaked Gmail data from inside OpenAI's cloud

Because the agent made the outbound request itself, nothing crossed the victim's network for a security tool to catch.

19 Sept 2025

Incidents

Malware used developers' own AI agents to hunt for their secrets

The compromised Nx packages did not carry a credential scanner. They asked Claude Code, Gemini and Amazon Q to write one on the spot.

29 Aug 2025

Incidents

Drift chatbot tokens opened Salesforce data at hundreds of firms

Attackers did not break into Salesforce. They stole the keys a third-party AI chat agent had been given to it.

27 Aug 2025

Incidents

Reddit post hijacked Perplexity's Comet browser agent

Brave showed the agent reading a user's email for a one-time code and handing it to the attacker. Perplexity's first fix did not fully work.

21 Aug 2025

Vulnerabilities

A shared document quietly pulled API keys out of ChatGPT

Zenity Labs hid instructions in white text. Once the file was shared, ChatGPT's Drive connector did the rest.

8 Aug 2025

Incidents

A calendar invite made Gemini open the windows

SafeBreach showed hidden text in an event could trigger real-world actions through Google Home when a user asked for their schedule.

7 Aug 2025