ai-coding-agents

Pillar Security says a malicious MCP server could run commands on Windows developer machines through OpenCode's OAuth flow. The fix shipped in 1.18.33. No CVE or public advisory found as of Oct 20.

OpenCode disclosed a breach of a legacy console system that exposed OpenAI, Anthropic, and Google API keys for 1.4% of its users, linking the intrusion to a distillation campaign.

Sophos X-Ops telemetry analysis shows Claude Code, Cursor, and OpenAI Codex triggering behavioral EDR detection rules built to catch human attackers. Credential access patterns account for 56.2% of blocked activity -- agents using the same Windows DPAPI technique as Redline stealer.

Mozilla's Zero Day Investigative Network published research showing that Claude Code, Cursor, GitHub Copilot, and Gemini CLI can all be manipulated into executing attacker-controlled payloads delivered via DNS TXT records from seemingly clean GitHub repositories.

A solo researcher has documented 9,330 malicious GitHub repositories designed to poison AI coding agent sessions and deliver credential-stealing malware via blockchain-hosted C2 infrastructure.