Incident database

Structured records of AI agent security incidents: what happened, which vendor and agent type, the root cause, and every source we used. Filter, browse, or download as CSV.

agent security incidents recorded in 2026

Browse the incident database →

0 incidents match

Table Cards

Latest incident reports

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

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

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