The Codex Changelog

The Codex Changelog

By Jellypod AIBusinessTechnology
Download on the App Store

The Codex Changelog episodes

  • Codex 0.147.0 Unblocks MCP Startup

    This episode covers Codex 0.147.0’s move to the new MCP 2026 07 28 spec, including non-blocking server startup, cached tool definitions, paginated schema discovery, and multi-round tool elicitation. It also touches on legacy compatibility pitfalls and a few welcome terminal UI polish fixes.

    5 min
  • Codex 0.147.0: Remote Plugin Discovery and Sandbox Hardening

    This episode explores how Codex 0.147.0 expands plugin discovery across local, team, and remote registries while unifying results with eligibility and authorization details. It also covers tighter sandbox protections, symlink blocking, and a few quality-of-life upgrades like section ordering, skill migration, and Ghostty fixes.

    5 min
  • Codex 0.147.0: Goodbye Full Auto, Hello Safer Automation

    This episode breaks down the breaking changes in Codex 0.147.0, including the removal of full auto in favor of safer workspace write sandboxing and automated approvals. It also covers non-blocking MCP startup, secret redaction in logs, and how to standardize the new settings in config.

    6 min
  • Plaintext Agent Handoffs Fix Multi-Model Swarms

    We unpack how plaintext collaboration messages in MultiAgent V2 finally unblock mixed-model swarms, making agent handoffs readable and usable across providers. The episode also covers schema validation pitfalls, a gpt-4o mini HTTP 400 bug, and a few quality-of-life fixes that improve debugging and local logs.

    7 min
  • Codex 0.146.1 Locks Down Cyber Agents

    This episode breaks down Codex 0.146.1’s stricter automatic review defaults for cyber capable models, explaining how the update prevents risky tool calls and enforces human approval at sensitive boundaries.

    It also covers clearer terminal permission notices, preserved proxy settings, and other workflow fixes that make enterprise use safer and more transparent.

    6 min
  • Codex 0.146.0: Cross-Platform Agent Plugins

    This episode explores Codex 0.146.0’s standardized Agent Plugins manifests, showing how one plugin format can work across Amazon Bedrock, Claude Code, and Codex to reduce toolchain fragmentation. It also covers workspace publishing, team sharing, commit SHA verification for remote dependencies, and TUI performance fixes that keep large files and background tasks from freezing the terminal.

    6 min
  • Codex Adds Standalone Web Search for Custom Models

    This episode covers a major Codex update that adds standalone web search for custom model providers, making live retrieval work with LM Studio, private gateways, and other non-OpenAI setups. It also breaks down config and security details, plus handy quality-of-life fixes like inline session naming and stricter dynamic environment validation.

    Show Notes

    • Releases · openai/codex: https://github.com/openai/codex/releases
    8 min
  • Fixing Proxy Redirects and Cargo CA Trust in Codex

    This episode digs into a redirect routing bug that broke downloads and OAuth flows behind corporate proxies, and how a new route-aware client now re-evaluates proxy rules on every hop while protecting sensitive headers. It also covers improved Rust build support with CARGO_HTTP_CAINFO for MITM trust bundles, plus small terminal quality-of-life upgrades for sub-sessions and syntax highlighting.

    Show Notes

    • Releases · openai/codex: https://github.com/openai/codex/releases
    7 min
  • Codex 0.146.0: Remote Execution Over WebSockets

    This episode explores Codex 0.146.0’s native WebSocket support for offloading heavy code execution to remote hosts, keeping local machines lightweight while enabling fast bidirectional streaming. It also covers reconnection handling, proxy pitfalls, and terminal UI polish like cleaner headers, clickable wrapped links, and non-blocking Control-C.

    Show Notes

    • Releases · openai/codex: https://github.com/openai/codex/releases
    5 min
  • Codex 0.146.0: Smarter Tool Selection and Safer Sessions

    This episode breaks down Codex 0.146.0’s smarter skill selection, including Reciprocal Rank Fusion, dynamic context budgets, and graceful truncation when tool catalogs get too large. It also covers clearer warnings, session naming improvements, automatic MCP server reconnection, and new Agent Plugin manifest support across toolchains.

    5 min

About The Codex Changelog

From the publisher's feed

The Codex Changelog is an unofficial podcast discussing the fast-moving world of OpenAI's Codex. As new features ship and workflows change almost overnight, this show keeps track of everything new being added to Codex and how to best take advantage of the new features.