The Codex Changelog

We break down the latest Codex CLI 0.153.0 updates, including account-scoped MCP approvals that prevent cross-account tool leakage, plus more reliable relative path launches on macOS. The episode also covers compressed rollouts, improved resume behavior, earlier usage warnings for Plus and Team users, and a small TUI config migration.

Show Notes

• Releases · openai/codex - GitHub: https://github.com/openai/codex/releases

We break down Codex 0.153.0’s experimental context mode, including the new new_context tool, smarter token budgeting, and which plans can actually enable it. The episode also covers quality-of-life upgrades like plugin marketplace commands, Vim mode fixes, auto recap controls, and stronger TUI reconnection handling.

This episode covers OpenAI Codex 0.152.1 fixes for Guardian authorization persistence during context compaction, plus faster Node REPL execution through model-level policy support. It also digs into terminal security hardening, cloud task origin isolation, and quality-of-life updates like actionable rate limit banners and Vim-style draft search.

We break down Codex CLI 0.152.0, where turn planning is now off by default to cut prompt overhead and token usage, with guidance on when to re-enable it for bigger refactors. The episode also covers MCP output limits, session-safe truncation, expanded server naming, Vim-mode polish, and longer shell command timeouts.

We break down Codex 0.151.0’s biggest workflow fixes, from per-repository plugin catalogs that isolate bad configs to configurable MCP startup grace periods and smarter model fallback handling.

Plus, we cover improved tool planning across mid-session model switches, new diagnostics, and permission profile fixes that keep development environments stable and predictable.

We unpack a subtle cross-platform sandbox bug in remote code execution, where path rules were being evaluated on the client OS instead of the remote executor. The episode also covers model-aware tool routing updates and improved preservation of structured MCP errors for more reliable integrations.

We break down Codex 0.151.0’s biggest agent-safety upgrades, including root-level token budgeting for subagents and middleware that can inspect, sanitize, or replace MCP tool output before it reaches the model. The episode also covers sandbox hardening fixes, permission profile preservation, Guardian cache invalidation, and startup grace period improvements.

This episode breaks down Codex CLI 0.149.0, highlighting codex queue for safely sending instructions to background agents without breaking workflow. It also covers new in-session directory commands and why explicit session UUIDs matter for reliable automation.

This episode breaks down Codex 0.150.0’s new cross-task mentions, showing how agents can message, query, and spawn subtasks without the usual copy-paste chaos. It also covers key security hardening for untrusted repos, Interrupt lifecycle hooks for cleanup, and quality-of-life upgrades like smarter task naming and copy controls.

We break down the latest Codex 0.148.0 upgrades, including non-blocking async hooks, cleaner process-tree termination, and new ways for hooks to call MCP tools automatically. The episode also covers improved conversation export, session forking and archiving, plus smoother OAuth recovery for long-running MCP workflows.