Agent DailyAgent Daily
releaseintermediate

[Release] openclaw/openclaw v2026.7.2-beta.6: openclaw 2026.7.2-beta.6

By vincentkocgithub
View original on github

OpenClaw v2026.7.2-beta.6 introduces major enhancements for state safety, durable message delivery, session management, and multi-platform support. Key features include crash-recoverable data protection with quarantine stores, conversation branching and rewinding, interactive MCP Apps with dashboards, structured questions/approvals across platforms, meeting integration with Teams/Zoom/Google Meet, Wear OS companion app, and guided local inference setup. The release also adds Claude Opus 5 support, native apps for macOS/Linux, enhanced memory management, and improved scheduling with dynamic cadence.

Key Points

  • Implement state safety with quarantine stores, crash-recoverable SQLite snapshots, and rollback-writer recovery to protect persisted data from database damage
  • Enable durable channel delivery across Telegram, Signal, Slack, QQBot, Twitch, and other platforms through shared ingress drain and dead-letter recovery mechanisms
  • Support session rewind and branching—fork conversations from individual messages, switch transcript branches across web/native apps, and preserve branch-safe queued sends
  • Host interactive MCP Apps with ticketed access, bound tools/resources, bounded context updates, and ability to pin them to durable dashboards
  • Deploy structured questions and approvals everywhere with option cards, push notifications, fair queuing, and Claude tool-request relay across web, channels, macOS, and native apps
  • Integrate real-time meetings with Teams, Zoom, and Google Meet with default-enabled plugins and durable transcript collection; add OpenAI/Gemini video support
  • Launch Wear OS companion with home-screen agent/session/model selection, realtime Talk controls, and audio-reactive playback
  • Provide guided setup for browser, Linux, and macOS with local-provider detection, downloadable models, and in-process llama.cpp/Gemma inference
  • Add Claude Opus 5, Kimi K3, and GPT Live realtime support with Platform API authentication; support local inference providers and Baseten Model API
  • Enhance security with channel allowlist fixes, session export containment, forged-marker bypass closure, and hardened secret redaction for exec/OAuth approvals

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