releaseintermediate
[Release] openclaw/openclaw v2026.3.1: openclaw 2026.3.1
By steipetegithub
View original on githubOpenClaw v2026.3.1 release introduces major enhancements across multiple platforms including adaptive thinking defaults for Claude 4.6, Kubernetes health check endpoints, expanded Android device capabilities (camera, notifications, sensors), Discord/Telegram session management improvements, and WebSocket-first OpenAI streaming. The release also adds localization support (German, zh-CN), Feishu document/chat tooling, LanceDB memory customization, and numerous reliability fixes for Android, Windows, LINE, and Slack integrations.
Key Points
- •Set Claude 4.6 models to use 'adaptive' thinking level by default while keeping other reasoning-capable models at 'low' unless explicitly configured
- •Add built-in HTTP health check endpoints (/health, /healthz, /ready, /readyz) for Docker/Kubernetes container probes with fallback routing to avoid shadowing existing handlers
- •Expand Android node capabilities with camera, device permissions, health monitoring, and notification actions (open/dismiss/reply) plus first-class node-tool actions
- •Replace Discord thread fixed TTL with inactivity-based lifecycle (idleHours default 24h) and optional hard maxAgeHours limit with /session commands for control
- •Add per-DM topic configuration for Telegram with allowlists, policies, and topic-aware authorization/debounce for messages, callbacks, commands, and reactions
- •Make OpenAI Responses WebSocket-first by default with SSE fallback and preserve server-side payload mutations (store + context_management) on WS path
- •Implement lightweight bootstrap mode for automation runs (--light-context for cron and agents.*.heartbeat.lightContext) to reduce context overhead
- •Replace ad-hoc subagent completion handoffs with typed internal completion events (task_completion) for consistent rendering across direct and queued announce paths
- •Add new optional 'diffs' plugin tool for read-only diff rendering from before/after text or unified patches with gateway viewer URLs
- •Support custom OpenAI baseUrl and embedding dimensions for LanceDB memory integration
Found this useful? Add it to a playbook for a step-by-step implementation guide.