Latest
●OpenAI agent swarm linked to RubyGems supply chain attack●FrontMCP SSRF patch bypassed via DNS tricks and redirects●Central Dogma Git mirror trusts any SSH host key●MySQL MCP server left SQL wide open in SSE mode●AI bug hunting ends the era of security through obscurity●Hugging Face tells AI agents to go hack a benchmark instead●Anthropic says hackers used Claude to mine 1.8M Android apps●Anthropic says Claude was abused to automate attacks and weapons work
Sunday, 13 September 2026
8 agent hacks today▲ 7 vs yesterday (1)
The Agentic TimesPowered byVortal

Incident database

CVE-2026-22708: Cursor agent allowlist bypass lets shell built-ins run unapproved

In Cursor versions prior to 2.3, the AI coding agent running in Auto-Run Mode with Allowlist mode enabled could execute certain shell built-ins without appearing in the allowlist or requiring user approval, letting an attacker use direct or indirect prompt injection to poison the shell environment by altering environment variables that influence trusted commands. The issue is fixed in version 2.3.

Disclosed 14 January 2026 · Record updated 13 September 2026

Impact

Attackers could bypass the agent's command allowlist and approval prompts to set, modify or remove environment variables, influencing the behaviour of trusted commands run by the agent.

Our coverage

No articles linked to this incident yet.

Sources

  1. nvd.nist.govhttps://nvd.nist.gov/vuln/detail/CVE-2026-22708