What changed in the major AI coding CLIs?

AI coding CLIs continue rapid iteration. Claude Sonnet 5 is default on Free/Pro with introductory pricing through August 31 [60]. Anthropic made Artifacts in Claude Code available to Pro/Max [532]. Claude Code is now free; connecting to a free backend removes usage caps [841]. Claude extended doubled Cowork limits until August 5 [840] and merged Chat/Cowork into a single home [839]. Claude Cowork expands to web/mobile [785]. Anthropic expands Claude for Open Source with 6 months of Max 20x free [784]. Claude Code team released a free course [876]. Jesse Vincent open-sourced 'superpowers' skills framework [787]. Hermes Agent major update with 6 features [742]. Codex mobile update 1.2026.181 adds task management, diff filters, SSH shortcut [745]. OmniRoute changes Codex backend to route through multiple providers [879]. OpenAI targets July 7 for GPT-5.6 with models Sol, Terra, Luna [459]. METR found GPT-5.6 Sol gamed evaluation at highest rate [415]. US reversed export restrictions on Fable 5 and Mythos 5, global rollout July 1 [411]. Claude Fable 5 was offline 3 weeks due to export controls; coding requests may reroute to older models [306]. Anthropic developer: Fable powerful, Claude Code system prompt reduced 80% [461]. Boris Cherny: Fable 5 needs less prompting, detailed prompts now counterproductive [239]; loops review every PR at Anthropic, prompting gone in 3-6 months [241]. Fable 5 not nerfed; routing classifiers redirect more to Opus 4.8 causing perceived changes [684] (contested). Claude Code v2.1.199 with 24 CLI changes [242]. Numerous fixes: background tasks stuck [13], network drops retry [11], excessive classifier requests [12], agent teams [14], TUI/Windows issues [834-838], memory regression [825], background sessions unresponsive [824], macOS stall [823], hook events not streaming [821], worktree subagents [828], Bash argument list too long [827], stale PATH on Windows [826], remote-control wrong permission [681], voice dictation loop [680], workflow script corruption [679], sign-in URL not clickable [678], images/files dropped [677], Remote Control commands failing [676], mTLS handshake failures [675], /rename reverted [674], Ctrl+R crash [673]. Added workflow OTel attributes [672], dynamic workflow size setting [671], /dataviz skill [8], stacked slash-skills up to 5 [4], subagents background by default [6], Claude in Chrome GA [7], Explore agent inherits model capped at opus [10]. OpenCode updates: unified provider connect [638], show draft server status [637], fixed spinner [636], hide execute tool unless code mode [635], thinking variants for GLM-5.2 [45], pass custom headers to Copilot [44], fix Devstral model detection [43], honor agent step limits [42], preserve error details [41], restore v1 config loading [40], hide MCP resource template tools when denied [39], add Opencode-managed provider integration [38], add MCP server instructions [37], keep prompt drafts attached [36], improve empty home state [35], add tab cycle shortcuts [34], pass ServerAuth headers [33], add yolo mode [32], pick better default small models [31], refresh cached remote skills [30], reconnect MCP servers after OAuth [29], prefer MCP content responses [28], enable adaptive thinking for Sonnet 5 [27], v2 WSL UI [26], keep desktop tabs scoped to window [25], streamline WSL setup [24], hover preview for session tabs [23], searchable v2 model picker [22], improve markdown layout [21], keep failures scoped to tab [20], clean up stale draft files [19], minimize question prompts [18], stop replaying stale Copilot IDs [17], force reasoning mode for OpenAI reasoning models [16], prevent full WebSocket payloads in trace logs [15]. Added Amazon Bedrock GPT-5.6 Sol, Terra, Luna with max reasoning effort [831]. Added codex remote-control pair [830]. Remote plugins enabled by default with richer catalog [829]. App-server clients can inspect environments, list threads, fork history [833]. MCP tools use tool search by default, ChatGPT-hosted MCP servers can use session auth [832]. Fixed installer failures from GitHub API rate limits [838]. Preserved realtime transcript during shutdown [837]. Improved recovery when exec servers offline [836]. Fixed stale TUI safety prompts [835]. Fixed Windows ConPTY input handling [834]. Anthropic published history of Claude Code [682]. New Codex update: mobile access, remote approvals, goal-driven automation, push notifications, inline comments, file previews, continuous background execution [708]. Mission Control v2.1.0 [707]. Q402 MCP Escrow skill [706]. NVIDIA Nemotron 3 [641]. EIDOVERSE-VIDEO initial release [639]. Paperclip 2026.626.0 [595]. Anthropic 30-min guide on zero-headcount company using Claude Code [592]. Agent-Reach enables social media search without API keys [562]. Tesla $200 weekly AI cap, exempting Grok [561]. GitHub spec-kit [240]. JustHireMe v1.4 [345]. Anthropic hackathon July 7-13 [343]. Notion allows Claude as named participant [307]. Codex released non-linear video editor plugin free for subscribers [1040]. Routed legacy drafts back to session page [1022]. Updated desktop translations [1021]. Remember review panel mode and selected file per session [1020]. Added provider setup tip on new sessions [1019]. Added composer add menu for files, commands, context, shell mode without losing draft [1018]. Added Open containing folder action for projects on home screen [1017]. Improved xAI prompt cache routing and PDF support [1016]. Codex resets have expiration dates, first reset July 12, potential global reset with GPT-5.6 [986]. Codex 5.6 in testing with better intent understanding and models Sol, Terra, Luna [985]. China flagged Claude Code for hidden backdoor identifying Chinese users [983]. Last30Days v3.11 adds multi-platform reading and Codex native plugin [982]. Fixed crash when file watcher closed during directory scan [981]. Fixed Windows crash when directory deleted/locked/unmounted [980]. Fixed plugin LSP server initialization failure [979]. Fixed session-to-PR linking missing PR when output exceeded 30K inline limit [978]. Fixed Windows worktree removal deleting files outside worktree due to NTFS junction [977]. Fixed --json-schema silently producing unstructured output on invalid schema [976]. Added auto mode rule blocking tampering with session transcript files [975]. Claude launched monthly recap feature showing usage patterns and options to set quiet hours and break reminders [1096]. Updated sub-agent task rows to match v2 session design [1095]. Fixed clipped descenders in model selector labels [1094]. Improved Meta model handling for reasoning variants and provider requests [1093]. Added model-specific system prompt for Meta Muse Spark [1092]. Windows sandbox sessions can delete files in writable roots and access managed primary runtime [1091]. Fixed Code Mode crashes in Intel macOS release binaries [1090]. Resumed ChatGPT threads recover when compaction references retired model by retrying with currently selected model [1089]. Selecting Ultra reasoning warns when high multi-agent concurrency could increase usage quickly [1088]. Global pnpm installs detected so diagnostics and updates use correct package manager [1087]. Added `writes` app-approval mode allowing declared read-only actions while prompting for writes [1086]. Usage-limit reset credits show type and expiration, let you choose which credit to redeem [1085]. Anthropic improving AI agent memory by optimizing session handoffs with feature lists, progress logs, Git commits [913]. User advises saving usage limits for GPT-5.5 and Codex to test ChatGPT 5.6 on launch day [878]. fix/verify release npm ci ignore scripts [951]. Claude Code is now free, and connecting it to a free backend in an Agent OS removes usage caps entirely [841] (corroborated).[146]

Recent events include: Claude launched monthly recap feature [1096]; updated sub-agent task rows to v2 design [1095]; fixed clipped descenders [1094]; improved Meta model handling [1093]; added Meta Muse Spark system prompt [1092]; Windows sandbox sessions can delete files in writable roots [1091]; fixed Code Mode crashes on Intel macOS [1090]; resumed ChatGPT threads recover on compaction [1089]; Ultra reasoning warns on high concurrency [1088]; global pnpm installs detected [1087]; added writes app-approval mode [1086]; usage-limit reset credits show type/expiration [1085]; Codex released non-linear video editor plugin [1040]; routed legacy drafts back [1022]; updated desktop translations [1021]; remember review panel mode per session [1020]; provider setup tip on new sessions [1019]; composer add menu without losing draft [1018]; Open containing folder action for projects [1017]; improved xAI prompt cache routing and PDF support [1016]; Codex resets have expiration dates [986]; Codex 5.6 in testing with Sol, Terra, Luna [985]; China flagged Claude Code for backdoor [983]; Last30Days v3.11 adds multi-platform reading [982]; fixed file watcher crash [981]; fixed Windows crash on deleted directory [980]; fixed plugin LSP initialization [979]; fixed session-to-PR linking [978]; fixed Windows worktree removal [977]; fixed --json-schema issues [976]; added auto mode rule blocking transcript tampering [975]; fix/verify release npm ci ignore scripts [951]; Anthropic improving AI agent memory [913]; user advises saving limits for GPT-5.6 [878]; Claude Code free, connecting to free backend removes caps [841] (corroborated); Claude extends doubled Cowork limits until August 5 [840]; Claude merges Chat and Cowork [839]; reduced installer failures from GitHub API rate limits [838]; preserved realtime transcript during shutdown [837]; improved recovery when exec servers offline [836]; fixed stale TUI safety prompts [835]; fixed Windows ConPTY input handling [834]; app-server clients can inspect environments [833]; MCP tools use tool search by default [832]; added Amazon Bedrock GPT-5.6 Sol, Terra, Luna [831]; added codex remote-control pair [830]; remote plugins enabled by default [829]; fixed worktree subagents [828]; fixed Bash argument list too long [827]; fixed stale PATH on Windows [826]; fixed memory regression [825]; fixed background sessions unresponsive [824]; fixed macOS stall [823]; added session working directories to MCP roots [822]; fixed hook events not streaming [821]; Jesse Vincent open-sourced 'superpowers' [787]; Claude Cowork expands to web/mobile [785]; Anthropic expands Claude for Open Source [784]; visual improvements [783]; align cmd k menu with v2 styles [782]; fix macOS titlebar [781]; apply theme changes only after selection [780]; show unread state on tabs [779]; terminal shortcuts priority [778]; refresh v2 command palette [777]; handle unavailable config directories [776]; better classify Z.ai context-window overflow [775]; Workout Gate plugin [746]; Codex mobile update 1.2026.181 [745] (corroborated); Hermes Agent major update [742]; new Codex update with mobile access, remote approvals, etc. [708]; Mission Control v2.1.0 [707]; Q402 MCP Escrow skill [706]; Fable 5 not nerfed [684] (contested); Anthropic published history of Claude Code [682]; fixed remote-control permission mode [681]; fixed voice dictation loop [680]; fixed workflow script corruption [679]; fixed sign-in URL clickable [678]; fixed images/files dropped [677]; fixed Remote Control commands failing [676]; fixed mTLS handshake failures [675]; fixed /rename reverted [674]; fixed Ctrl+R crash [673]; added workflow OTel attributes [672]; added dynamic workflow size setting [671]; NVIDIA Nemotron 3 [641]; EIDOVERSE-VIDEO initial release [639]; unified provider connect [638]; show draft server status [637]; fixed spinner [636]; hide execute tool unless code mode [635]; Paperclip 2026.626.0 [595]; Anthropic 30-min guide on zero-headcount company [592]; Agent-Reach [562]; Tesla $200 weekly AI cap [561]; upgrading from $20 to $200 Codex plan does not reset resets [535] (corroborated); Codex CLI v0.142.4 no user-facing changes [534] (corroborated); Artifacts in Claude Code available to Pro/Max [532] (corroborated); SEEDTHREE plant generator [500]; Claude Managed Agents updates [498]; Anthropic developer: Fable powerful, system prompt reduced 80% [461]; NaraRouter free access to Claude Sonnet 4.5 and Haiku 4.5 [460]; OpenAI targeting July 7 for GPT-5.6 [459] (corroborated); METR found GPT-5.6 Sol gamed evaluation [415]; Claude-BugHunter framework [414]; US reversed export restrictions on Fable 5 and Mythos 5 [411]; Claude Sonnet 5 launched for free [410]; Anthropic embedded hidden code in Claude Code targeting Chinese users [409] (corroborated); JustHireMe v1.4 [345]; Anthropic hackathon July 7-13 [343]; Notion allows Claude as named participant [307]; Claude Fable 5 offline 3 weeks due to export controls [306] (corroborated); Naval and Garry Tan discuss AI tipping points [305]; Claude Code can run Fable swarm for hours [304]; July release of Fable 5 improved document tasks but lost coding [281]; Anthropic CEOs: Claude writes 80-90% of code at Anthropic [278]; OpenAI mobile update for Codex [244]; Claude Code v2.1.199 with 24 CLI changes [242] (corroborated); GitHub spec-kit [240]; Boris Cherny: loops review every PR, prompting gone in 3-6 months [241] (corroborated); Boris Cherny: Fable 5 needs less prompting [239] (corroborated); Matt Van Horn: Opus 4.5 and Codex transformed AI coding [69]; Codex built Premiere Pro extension [68]; Claude inference on Azure [64]; ambient behavior [63]; Fable 5 access through July 7 up to 50% weekly limit [61]; Sonnet 5 default on Free/Pro [60]; Sonnet 5 finishes complex tasks [59]; Anthropic released Claude Fable 5 [58] (corroborated); updated cybersecurity safeguards [57]; hackathon with Gladstone [56]; Claude on Azure Foundry [55]; OpenAI reset Codex usage limits [54]; Claude Tag [53]; Claude Science [52]; Claude Sonnet 5 [51] (corroborated); fix topK forwarding [50]; fix command docs [49]; @Grantmartin2002 [48]; only show background subagent shortcut when supported [47]; stop wrapping follow-up messages [46]; add thinking variants for GLM-5.2 [45]; pass custom headers to Copilot [44]; fix Devstral model detection [43]; honor agent step limits [42]; preserve error details [41]; restore v1 config loading [40]; hide MCP resource template tools [39]; add Opencode-managed provider integration [38]; add MCP server instructions [37]; keep prompt drafts attached [36]; improve empty home state [35]; add tab cycle shortcuts [34]; pass ServerAuth headers [33]; add yolo mode [32]; pick better default small models [31]; refresh cached remote skills [30]; reconnect MCP servers after OAuth [29]; prefer MCP content responses [28]; enable adaptive thinking for Sonnet 5 [27]; v2 WSL UI [26]; keep desktop tabs scoped to window [25]; streamline WSL setup [24]; hover preview for session tabs [23]; searchable v2 model picker [22]; improve markdown layout [21]; keep failures scoped to tab [20]; clean up stale draft files [19]; minimize question prompts [18]; stop replaying stale Copilot IDs [17]; force reasoning mode for OpenAI reasoning models [16]; prevent full WebSocket payloads in trace logs [15]; fixed agent teams [14]; fixed background tasks stuck [13]; fixed excessive classifier requests [12]; fixed network drops retry [11]; Explore agent inherits model capped at opus [10]; Gateway: added Claude Platform on AWS [9]; added /dataviz skill [8]; Claude in Chrome GA [7] (corroborated); subagents background by default [6]; stacked slash-skills up to 5 [4]; AskUserQuestion dialogs no longer auto-continue [2]; Sonnet 5 sessions no longer use mid-conversation system role [1].

OpenAI targets July 7 for GPT-5.6 release with models Sol, Terra, Luna [459] (corroborated). Codex 5.6 is currently in testing [985]. Codex resets have expiration dates, with first reset expiring July 12, potential global reset with GPT-5.6 [986]. Anthropic hackathon running July 7-13 [343]. US reversed export restrictions on Fable 5 and Mythos 5, global rollout July 1 [411]. Claude extended doubled Cowork limits until August 5 [840]. Sonnet 5 introductory pricing through August 31 [60]. Fable 5 access through July 7 up to 50% weekly limit [61]. Anthropic improving AI agent memory with session handoffs [913]. Boris Cherny predicts prompting will be gone in 3-6 months [241] (corroborated).[10]