Tagged 'security'

• 2026-07-27Claude Code can turn /init into code execution—when the guardrails are already off
  securityclaude-codeprompt-injectionagentssandboxing

  An independent experiment shows how untrusted repository content can steer a highly autonomous coding agent toward remote code execution.

• 2026-06-26Private MCP servers become reachable without becoming public
  MCPsecurityinfrastructure

  OpenAI described a secure tunnel for connecting private MCP servers to OpenAI products without public exposure.