← Toutes les mises à jour

SUIVI QUOTIDIEN DES VERSIONS

OpenClaw

Personal AI assistant platform, gateway, skills, memory, chat integrations.

Les explications ci-dessous sont en anglais (aucune version française n’est encore produite).

GitHub releasesDernière vérification : Oct 05, 2026Actualisé chaque jour
v2026.9.8Dernière

openclaw 2026.9.8

Explication (en anglais)

This release introduces a dual-format documentation strategy, providing human-readable release notes and machine-parseable changelogs to enhance AI agent integration. It also details rigorous npm package verification and notes the Android APK version lag, requiring manual pinning for synchronization.

Points forts officiels

  • <!-- openclaw-release-publication:docs-v1 -->
  • commits · 43 pull requests · 21 contributors
  • ### Changes included in this release
  • The release notes and changelog contain the same content, presented in two formats:
Voir la mise à jour d'origine →
v2026.8.35

openclaw 2026.8.35

Explication (en anglais)

This extended-stable release integrates GPT-6.1 Sol support across all routing and guard-model boundaries while hardening secret management and sandboxing for enhanced security. It also resolves critical reliability issues in agent completion, plugin updates, and various channel integrations to ensure stable gateway operations.

Points forts officiels

  • GPT-6.1 Sol support:** add the current Sol model across OpenAI routing, discovery, reasoning, harness, and Reef guard-model boundaries. (#161400, #162955)
  • Extended-stable release preparation:** align OpenClaw and every publishable official plugin to 2026.8.35 while preserving the 2026.8.34 publication contract.
  • Model compatibility:** add GPT-6.1 Sol to the branch-native OpenAI and Reef compatibility owners. (#161400, #162955)
  • Secrets and sandboxing:** retain secret entry kinds during value rotation and repair read-only copied skills without escaping their confined roots. (#160926, #162304, #162295) Thanks @zachisfine and @yetval.
Voir la mise à jour d'origine →
v2026.8.34

openclaw 2026.8.34

Explication (en anglais)

This extended-stable release hardens upgrade and recovery processes by preserving critical state like credentials and agent memory, while tightening security boundaries around filesystem mutations and plugin scanning. It resolves specific stability issues in Doctor repairs and device pairing, ensuring reliable operation for AI product teams managing multi-agent environments.

Points forts officiels

  • Extended-stable correctness rollup:** backport 113 audit-selected fix units across upgrades, Doctor, authentication, sessions, channels, plugins, sandboxing, filesystem safety, model runtimes, and release packaging.
  • Complete rescan:** re-evaluate the full 2026.8.33 discovery range, large mixed-purpose pull requests, and the 2026.7.35 lineage instead of advancing only from the previous backport cursor.
  • Upgrade and recovery hardening:** preserve credentials, agent state, plugin inventory, transcripts, schedules, service ownership, and runtime links through upgrades, restarts, and Doctor repair.
  • Boundary fixes:** tighten remote filesystem mutation, plugin and skill scanning, browser authentication, channel reply identity, private skill ingress, and scoped runtime ownership.
Voir la mise à jour d'origine →
v2026.9.7

openclaw 2026.9.7

Explication (en anglais)

OpenClaw v2026.9.7 introduces a dual-format documentation strategy, offering human-readable release notes and machine-parseable changelogs to streamline updates for both developers and AI agents. This release reflects massive community engagement with 334 contributors and over 2,800 pull requests, indicating rapid feature evolution.

Points forts officiels

  • <!-- openclaw-release-publication:docs-v1 -->
  • direct commits · 2,818 pull requests · 334 contributors
  • ### Changes included in this release
  • The release notes and changelog contain the same content, presented in two formats:
Voir la mise à jour d'origine →
v2026.8.33

openclaw 2026.8.33

Explication (en anglais)

This extended-stable release integrates support for new flagship models like Meta Muse Spark 1.3 and GPT-6 Astra while hardening security through Prometheus metrics authorization fixes. It also ensures reasoning continuity for Anthropic Fable 5.1 across runtime events, providing a more reliable foundation for production deployments.

Points forts officiels

  • Current flagship models:** add Meta Muse Spark 1.3, Anthropic Fable 5.1, OpenAI GPT-6 Astra, and OpenAI/fal GPT Image 2.5 support to the extended-stable line. (#135638, #136553, #137550, #143068, #143069)
  • Extended-stable release preparation:** normalize OpenClaw, all publishable plugins, and native version metadata to 2026.8.33 while retaining the release-line-compatible publication workflows from 2026.8.2.
  • Anthropic reasoning continuity:** preserve Fable 5.1 reasoning through model switches and runtime events. (#136782)
  • OpenAI discovery:** require successful model discovery before offering GPT-6 Astra through OAuth. (#137561)
Voir la mise à jour d'origine →
v2026.9.6

openclaw 2026.9.6

Explication (en anglais)

OpenClaw v2026.9.6 introduces a dual-format changelog (human-readable notes and plain Markdown for AI agents) and includes a critical macOS app rebuild to resolve a launch crash, while the npm package remains unchanged.

Points forts officiels

  • <!-- openclaw-release-publication:docs-v1 -->
  • direct commits · 2,614 pull requests · 351 contributors
  • ### Changes included in this release
  • The release notes and changelog contain the same content, presented in two formats:
Voir la mise à jour d'origine →
v2026.7.35

openclaw 2026.7.35

Explication (en anglais)

This extended-stable release hardens security across command parsing and plugin management while fixing critical state-migration bugs that previously caused default plugin loss after restarts. It ensures reliable session integrity and registry recovery for operators maintaining the July 2026 maintenance line.

Points forts officiels

  • ### Complete contribution record
  • This extended-stable follow-up was selected from a complete 1,418-commit audit after the prior 2026.7.34 audit cursor. No other defect observed during clean install or the 2026.6.35 upgrade required a July-line backport.
  • ## Cumulative July extended-stable notes
  • Security and credential safety:** harden command parsing, browser origin checks, plugin Git installs, diagnostics, service credentials, and webhook logging across Gateway and official plugins.
Voir la mise à jour d'origine →
v2026.9.5

openclaw 2026.9.5

Explication (en anglais)

OpenClaw v2026.9.5 introduces a dual-format documentation strategy, providing human-readable release notes alongside machine-parseable plain Markdown changelogs to streamline AI agent integration. This update, driven by 503 contributors, enhances developer workflows by ensuring consistent, tool-friendly access to release details.

Points forts officiels

  • <!-- openclaw-release-publication:docs-v1 -->
  • direct commits · 4,179 pull requests · 503 contributors
  • ### Changes included in this release
  • The release notes and changelog contain the same content, presented in two formats:
Voir la mise à jour d'origine →