What changed in legal AI workflows

OpenAI's June 16 Codex changelog says Computer Use, the Codex Chrome extension, Memories, and the Chronicle research preview are rolling out in the EEA, UK, and Switzerland, with Memories off by default in those regions. Claude Code 2.1.179 fixes mid-stream connection drops, large Linux sandbox read-rule glob expansion, remote-session background-task status, subagent transcript and focus issues, and remote plugin-loading performance; the allowlisted Claude Code Action also bumped to 2.1.179.

OpenAI's June 15 Codex CLI 0.140.0 changelog adds `/usage` token-activity views, permanent session deletion, Claude Code import, unified file/plugin/skill mentions, managed Bedrock API-key authentication, encrypted CLI/MCP OAuth credential storage, MCP startup and OAuth reliability fixes, and remote plugin installation fixes. Claude Code 2.1.178 adds parameter-scoped permission rules, nested `.claude/skills` loading, closest-directory project settings precedence, auto-mode review for subagent spawns, and fixes for MCP subagent `disallowedTools`.

Allowlisted Claude Code Action commits fixed two `claude_args` parsing gaps: unquoted scoped Bash rules such as `Bash(gh:*)` could collapse to broader Bash access, and MCP server installation could diverge from the SDK's parsed `--allowedTools` grant when multiple values or commented lines were present.

OpenAI's June 11 Codex app 26.609 changelog adds Developer mode for Browser use through Chrome DevTools Protocol access, `/init` in the app composer, Computer Use expansion and Windows per-app access controls, usage-limit guidance, plugin management improvements, and a fix so scheduled automations honor the selected approval mode. Allowlisted Claude Code Action commits also bump Claude Code to 2.1.175, preserve inherited auth environment variables when inputs are empty, and stop SDK workflows from hanging after a terminal result message.

OpenAI's June 4 API changelog adds moderation scores to Responses and Chat Completions generation requests, while Claude Code 2.1.163 adds managed minimum and maximum version settings, `/plugin list`, hook continuation context, MCP session ID continuity, and multiple permission, background-session, and Windows reliability fixes; Claude Code 2.1.165 is a reliability release.

OpenAI's June 2 Codex changelog adds Sites preview in the Codex app, iOS Codex controls for Face ID or passcode lock, Queue versus Steer follow-up defaults, Windows SSH connections, `/side <prompt>`, and host/task reliability fixes. The OpenAI API changelog also moves eligible container sessions to per-minute billing with a five-minute minimum, while an allowlisted Claude Code Action commit adds workload identity federation inputs to the base action and requires GitHub OIDC write permission for OIDC token exchange.

OpenAI's Codex 0.132.0 changelog adds Python SDK authentication, simpler text-only turn APIs, structured output support for resumed exec sessions, standard Codex auth for remote executors, and app-server image-fidelity preservation; the OpenAI API changelog also introduced Secure MCP Tunnel for account-led enterprise access to private MCP servers.

Codex CLI 0.131.0 adds richer TUI controls, unified file/plugin/skill mentions, plugin marketplace commands, remote-control support, the `openai-codex` Python SDK package, and `codex doctor`; Claude Code 2.1.144 adds background-session resume support and fixes plugin, MCP, headless, remote-login, and background-agent reliability issues.

OpenAI's Codex changelog adds mobile remote access through a connected Mac plus hooks, access-token, and enterprise setup guidance, while Claude Code 2.1.143 tightens plugin, worktree, hook, background-session, and Windows behavior.

Codex Action now calls PR-controlled project instruction files untrusted input, while Claude Code Action tightened tag-mode scoping so only the trigger comment or issue body acts as instructions.

Claude Code 2.1.141 adds hook terminal-sequence output, HTTPS plugin cloning, workspace-scoped federation, directory-scoped agent listing, recent-session feedback bundles, and background-agent permission fixes.

Claude Code 2.1.139 adds agent view, `/goal`, direct hook argument execution, PostToolUse continuation, MCP project-directory environment support, and subagent telemetry headers.

OpenAI expanded Codex Auto-review documentation and published Codex cookbook examples for iterative repair and agent-improvement loops using traces and evals.

GitHub's MCP Server gained Projects tooling, HTTP server mode, and permission-aware tool filtering, while Agentic Workflows entered technical preview for AI agents inside GitHub Actions.

GitHub MCP Server secret scanning is generally available for repositories with GitHub Secret Protection, giving MCP-compatible agents and IDEs a pre-commit and pre-PR credential check.

Research guidance now separates model knowledge from current source access and requires jurisdiction, date, treatment, and MCP tool-scope checks before legal reliance.

OpenAI docs now emphasize current GPT-5.5/GPT-5.4-era model guidance, updated ChatGPT plan naming, and Responses API plus built-in tools as the starting point for new API workflows.

Claude Code added worktree base controls, effort-aware hooks, managed sandbox settings, and fixes for MCP OAuth, subagent skill discovery, and concurrent sessions.

Codex docs now emphasize recurring app automations, linked-repo worktrees, browser permissions, subagents, plugins, and CLI sandbox or approval profile controls.

OpenAI released Realtime 2, streaming translation and transcription updates, and the OpenAI Developers plugin for Codex.

Claude Code GitHub Actions supports scheduled and event-driven coding workflows, including skills and prompt-driven automation that can help maintain tutorial quality.

We are adding a source-backed automation workflow that researches Codex and Claude changes, proposes tutorial updates, and hands approved changes to translation agents.