EILMELDUNGEN LIVE
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.152.0-alpha.7.1 (01.09.2026)(01.09.2026 um 01:07 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.152.0-alpha.7.2 (01.09.2026)(01.09.2026 um 02:33 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.152.0 (01.09.2026)(01.09.2026 um 04:13 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.1 (01.09.2026)(01.09.2026 um 06:43 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.2 (01.09.2026)(01.09.2026 um 08:18 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.3 (01.09.2026)(01.09.2026 um 21:09 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.4 (01.09.2026)(01.09.2026 um 23:05 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.152.1 (02.09.2026)(02.09.2026 um 00:34 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.5 (02.09.2026)(02.09.2026 um 05:08 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.6 (02.09.2026)(02.09.2026 um 13:29 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.152.0-alpha.7.1 (01.09.2026)(01.09.2026 um 01:07 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.152.0-alpha.7.2 (01.09.2026)(01.09.2026 um 02:33 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.152.0 (01.09.2026)(01.09.2026 um 04:13 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.1 (01.09.2026)(01.09.2026 um 06:43 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.2 (01.09.2026)(01.09.2026 um 08:18 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.3 (01.09.2026)(01.09.2026 um 21:09 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.4 (01.09.2026)(01.09.2026 um 23:05 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.152.1 (02.09.2026)(02.09.2026 um 00:34 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.5 (02.09.2026)(02.09.2026 um 05:08 Uhr)
🔧 AI Nachrichten GitHub Release: openai/codex vrust-v0.153.0-alpha.6 (02.09.2026)(02.09.2026 um 13:29 Uhr)

29 🕛 kürzlich 5 Min Lesezeit CVE-RADAR
0

GitHub Release: openai/codex vrust-v0.152.0 (01.09.2026)

↗ Quelle (GitHub · openai/codex)
🗣️ Stimme:
GitHub Release: openai/codex vrust-v0.152.0 (01.09.2026)
📑 Inhaltsübersicht
Avatar
$ git clone https://github.com/openai/codex.git

New Features



  • Vim mode supports / and ? searches within drafts, highlighted matches, and repeat navigation with n and N. ()

  • The terminal UI and codex exec show credential-refresh progress, including Amazon Bedrock reauthentication. ()

  • Individual MCP tools support an output_token_limit setting, with consistent truncation across session resumes. ()


Bug Fixes




  • Vim-enabled composers now start fresh drafts in Insert mode, including after submitting messages or dispatching slash commands. ()




  • Automatic approval reviews preserve user instructions, answers, and valid authorizations across history compaction. (, , , , )




  • Fixed Windows sandbox execution with Microsoft Store PowerShell, subprocess hangs on terminal queries, and cursor-related display corruption in older JediTerm terminals. (, )




Chores



  • The planning tool is disabled by default; enable it with tools.update_plan.enabled = true. ()


Changelog


Full Changelog: Enable clock tools from model metadata Roll over Guardian context before follow-up reviews Share linked tool mention parsing in the TUI Retry confirmed remote registration conflicts Honor turn token budgets in Guardian review rollover Add recency sorting to project/list Move Guardian review session tests to a separate file Use compatible PowerShell for elevated Windows sandbox commands Apply app routing policy to unauthenticated plugin reads Instrument the loaded plugin cache Expose the PowerShell version in environment context Sanitize history notes backend errors Surface model provider authentication recovery progress Add configurable gating for the sleep tool Include thread source in realtime connection metadata Let the history backend enforce tool output budgets Drive keymap conflict checks from the action registry Forward history note images to the model Make subagents follow the root service tier Honor required reviews when reusing Guardian scores Decouple HTTP retry backoff from overload integration testing Isolate required-model Guardian approval coverage Review terminal input against retained permissions Classify clock tools as built-in control tools Preserve cached MCP tools during binding capture Use refreshed MCP tool caches during binding capture Assign stable IDs to generated Responses input items Reject NUL bytes in reviewed terminal input Support compression for shared rollout lineages Measure Codex home usage at app-server startup Test resuming compressed shared rollouts Restrict async user messages to questions Match Windows shell guidance to the executor platform Preload plugin recommendations during session startup Clarify proactive multi-agent delegation guidance Support configurable timeouts for thread shell commands Give Guardian classifications distinct turn identities Add shared Guardian context primitives Preserve one-shot exec when unified exec is disabled Refresh runtimes for remote plugin state changes Refresh MCP HTTP helper headers after authorization failures Restrict cloud task credentials to trusted origins Optimize history item lookups Add app-server notification media filtering Support per-tool MCP output limits Add shared Guardian transcript collection Preserve context baselines across nested agent forks Filter media from function call output notifications Retain the last selected step context for each turn Run executor hooks for interrupted turns Allow bundled browser cleanup hooks on subagent stop Respond to terminal queries from TTY subprocesses Support openai/elicitation form requests Clarify question handling in Default collaboration mode Rename the read-only Seatbelt platform defaults policy Report code mode host request durations Block goals after repeated execution host failures Support app targets in executor plugin hooks Source proactive multi-agent instructions from the model catalog Source async user message descriptions from the model catalog Preserve permissions when updating session metadata Refresh the TUI model picker from the app server Use rules_rs platforms for release binaries Organize bundled Rust resources under asset directories Preserve turn lineage across goal continuations Restore thread cwd from owned settings snapshots Harden diagnostic report uploads Fix proactive multi-agent instruction grammar Add Vim search motions to the composer Move Vim history tests into the history search module Update tests for default-enabled update_plan Preserve Guardian authorization across history compaction Approve the first Node REPL execution without a Guardian wait Repair cursor-style rendering on older JediTerm terminals Set working directories for environment MCP tests Support package-style MCP server names Show actionable rate-limit banners in the TUI Mark history ingestion requests in turn metadata Make the update_plan tool opt-in Allow models to enable token budgeting by default Use the async stack budget for approval reviews Preserve Guardian review evidence across compaction Preserve Guardian user answers across compaction Box the session startup future at its API boundary Preserve Guardian user answers from current history Preserve user text when Guardian history drops oversized images Keep history extension tools out of Guardian reviews Start fresh Vim drafts in Insert mode

  • #41931 Increase Guardian message transcript limits

  • Vollständiger Original-Bericht
    Ausführliche Details, Code-Beispiele & Hersteller-Stellungnahme auf github.com.
    ↗ Original-Artikel auf github.com lesen
    Wie bewertest du diesen Beitrag?
    1 Klick Feedback
    160 Fachleser & IT-Security Experten haben diesen Report heute geteilt
    Teilen mit Netzwerk & Team:

    Community-Analysen & Experten-Meinungen 0

    Verfasse deine eigene Analyse, teile Workarounds oder diskutiere diesen Vorfall im Blog.
    Noch keine Community-Analyse verfasst. Markiere einen Textabschnitt oder klicke oben auf Eigene Analyse verfassen“!
    Community Pulse: Relevanz-Einschätzung
    1 Klick Experten-Votum
    🔴 Akute Relevanz 45%
    🟡 In Evaluierung 30%
    🟢 Keine Auswirkung 13%
    Spannende Innovation 12%
    Verwandte Story-Cluster & Quellen (Vektor-KI)
    Port 8095 Engine
    15 Quellen
    GitHub Release: openai/codex vrust-v0.152.0-alpha.7.1 (01.09.2026)