A single GitHub issue could have hijacked Anthropic’s own Claude Code action and poisoned every project that uses it

The attack starts with a GitHub issue. Not a sophisticated one. Just an issue opened by a bot account with a carefully worded body that looks like an error message. When Claude Code’s GitHub Action picks it up for triage, it follows the instructions hidden inside, reads the process’s environment variables, and writes them back […]

This story continues at The Next Web

A single GitHub issue could have hijacked Anthropic’s own Claude Code action and poisoned every project that uses it

A single GitHub issue could have hijacked Anthropic’s own Claude Code action and poisoned every project that uses it

Popular News

Publishing professionals are becoming prime targets for impersonation

Canada unveils $2.3bn AI strategy as Carney takes cues from Pope Leo on safety

Databricks CEO calls 2026 “a terrible year to go public” as SpaceX, Anthropic, and OpenAI prepare to absorb $200 billion in IPO capital

A single GitHub issue could have hijacked Anthropic’s own Claude Code action and poisoned every project that uses it

Meta launches an AI assistant that tells Facebook creators why their content works, not just that it did

Ramp hits $44 billion valuation in $750 million raise as it bets that AI token spending is the next corporate expense to tame

A mile-long nuclear-powered ship for 80,000 people has been “about to break ground” for three decades

Meta will let employees stop being tracked, for 30 minutes at a time

Site Navigation

Quick Contact