[Release] openclaw/openclaw v2026.6.5-beta.1: openclaw 2026.6.5-beta.1

OpenClaw v2026.6.5-beta.1 is a major release focusing on content safety, provider reliability, and platform stability. Key improvements include QQBot reasoning stripping to prevent model thinking leakage, MCP tool result coercion to prevent API errors, Anthropic extended-thinking session recovery, Parallel web search bundling, and enhanced durability for auth/plugin state. The release addresses 50+ bug fixes across channels, providers, agents, and infrastructure with improved test isolation and CI/CD reliability.

Key Points

• •QQBot now strips model reasoning/thinking scaffolding before delivery to prevent raw `<thinking>` content from leaking into channel replies
• •MCP tool results coerce non-text/image blocks at materialization boundary to prevent Anthropic 400s and poisoned session history
• •Anthropic extended-thinking sessions recover after prompt-cache expiry or Gateway restart by waiting for `message_start` events
• •Parallel is now a bundled `web_search` provider with `PARALLEL_API_KEY` discovery, guarded endpoints, and cache-safe session IDs
• •Google Vertex ADC users get static catalog rows and runtime model resolution with improved single-provider cooldown recovery
• •Matrix channels support voice-note preflight and thread-aware read/reply behavior with QA coverage
• •Auth profiles now live in SQLite with durable plugin install records and prerelease fallback integrity checks
• •macOS node mode no longer silently self-reconnects away from healthy direct Gateway sessions, reducing session churn
• •Cron legacy JSON stores migrate during doctor preflight; service env placeholders no longer mask state-dir secrets
• •Test isolation improved across provider, media, auth, cron, task, session, sandbox, Gateway, and Codex fixtures to reduce cross-test leakage

Workflow Diagram

Concepts