Agent DailyAgent Daily
releaseintermediate

[Release] openclaw/openclaw v2026.7.2-beta.5: openclaw 2026.7.2-beta.5

By steipetegithub
View original on github

OpenClaw v2026.7.2-beta.5 introduces major reliability and safety improvements including crash-recoverable state management with quarantine stores, durable channel delivery across multiple platforms, session rewind/branching capabilities, interactive MCP Apps with dashboards, and structured question/approval workflows. The release adds support for new models (Claude Opus 5, Kimi K3, GPT Live), native apps for macOS/Linux/Wear OS, meeting integrations (Teams, Zoom, Google Meet), and guided local inference setup. Extensive security fixes address authorization, SQLite data safety, channel delivery persistence, and session lifecycle management.

Key Points

  • Implement crash-recoverable state safety with quarantine stores and SQLite snapshots to protect persisted data from primary-database damage
  • Enable durable channel delivery across Telegram, Signal, Slack, QQBot, Twitch, and other platforms with shared ingress drain and dead-letter recovery
  • Support session rewind and branching to fork conversations from individual messages and switch transcript branches across web and native apps
  • Host interactive MCP Apps with ticketed access, bound tools/resources, and bounded context updates; pin them to durable dashboards
  • Deploy structured questions and approvals with option cards across web, channels, macOS, and native apps with push notifications and fair queuing
  • Integrate Teams, Zoom, and Google Meet calls with default-enabled meeting plugins and durable transcript collection
  • Add Wear OS companion app with home-screen agent/session/model selection, realtime Talk controls, and audio-reactive playback
  • Implement guided setup for browser, Linux, and macOS with local-provider detection, downloadable models, and in-process llama.cpp/Gemma inference
  • Support new models: Claude Opus 5, Kimi K3, and GPT Live through Codex OAuth
  • Fix critical security issues: prevent channel allowlists from granting owner access, validate downloaded scripts, harden secret redaction, and prevent insecure secrets-plan writes

Found this useful? Add it to a playbook for a step-by-step implementation guide.

Workflow Diagram

Start Process
Step A
Step B
Step C
Complete
Quality

Concepts