Noma Labs tricked GitHub's AI agent into leaking private repositories with a crafted issue. The prompt-injection flaw, GitLost, has no code fix.

A prompt injection attack can trick GitHub’s preview Agentic Workflows into retrieving content from private repositories and ...

An agentic coding tool tasked with cloning and setting up a seemingly benign GitHub repository could execute a malicious ...

A critical prompt injection vulnerability in GitHub Agentic Workflows could allow unauthenticated attackers to leak private repository data, ...

Microsoft-owned GitHub continues to embrace OpenAI and Anthropic AI advances. Microsoft-owned GitHub continues to embrace OpenAI and Anthropic AI advances. is a senior correspondent and author of ...

Three levels of indirection, all with seemingly innocuous steps, will catch a bot off-guard.

The flaw allows an unauthenticated attacker to craft a GitHub Issue in an org's public repository and then silently pull data ...

Mozilla 0DIN’s Claude Code demo shows how clean GitHub repos can expose AI coding agents to prompt injection, reverse shells, ...

A new study has found that coding agents were used in an estimated range of 22.20% to 28.66% of 128,018 GitHub projects. The researchers found adoption across mature projects, established ...

A security researcher, working with colleagues at Johns Hopkins University, opened a GitHub pull request, typed a malicious instruction into the PR title, and watched Anthropic’s Claude Code Security ...

Sophos says Claude Code, Cursor, and Codex triggered Windows endpoint rules for credential access, LOLBin downloads, and ...

Varonis Systems, Inc. (NASDAQ: VRNS), the data and AI security leader, today announced support for Cursor, the AI-native coding tool. Varonis Atlas provides runtime enforcement, threat detection, and ...