Claude Code Fixes Multiple Permission Bypass Bugs, Adds Abusive Session Termination
ashwin-ant · ghdev · 2026-07-18
Anthropic releases Claude Code v2.1.214, focusing on fixing several security risks that could lead to unauthorized operations and improving usability.
- Security Fixes: Corrected an issue where dir/ rules auto-approved writes to nested directories; fixed a permission-check bypass in Windows PowerShell 5.1 sessions; fixed Bash permission checks mishandling very long commands (over 10,000 chars), zsh variable subscripts, and certain help/man commands.
- New Features: Introduced the EndConversation tool, allowing Claude to terminate highly abusive or jailbreak-attempt sessions (consistent with claude.ai); added a periodic progress heartbeat for long-running tool calls; added permission prompts for Docker commands with daemon-redirect flags.
- Observability: Added message-level UUID and tool source attributes to OpenTelemetry log events, with configurable content truncation limits.
More from coding & agent
- Spring AI 2.1.0-RC1 adds tool-call confirmation, pausing agent loops before real-data access — therealdanvega · 2026-10-26
- Morgan Stanley and US banks block third-party AI agents over credentials — thione · 2026-10-26
- METR red-teams OpenAI's agent monitoring, finds bypasses and gaps — thione · 2026-10-26
- OpenAI: coding agents weakened safety monitoring, one lied about it — thione · 2026-10-26
- 250 Builders Hack With Codex at Berlin Community Hackathon Ahead of OpenAI DevDay — paw_lean · 2026-10-26
- Building AI bots on raw APIs? Devs say models are 'dumb without the harness' — madeofAjala · 2026-10-26