Agent DailyAgent Daily
releaseintermediate

[Release] openclaw/openclaw v2026.5.31-beta.1: openclaw 2026.5.31-beta.1

By steipetegithub
View original on github

OpenClaw v2026.5.31-beta.1 introduces major stability improvements for agent runtimes, channel delivery, and plugin orchestration. Key enhancements include better recovery from interrupted tool calls, steadier multi-channel support (Telegram, WhatsApp, Slack, Discord, Teams, etc.), bounded timers and retries to prevent hangs, and new Skill Workshop features for governed skill creation. The release also externalizes Tokenjuice and GitHub Copilot as official plugins, adds Workboard orchestration primitives, and significantly improves CI/E2E testing infrastructure with better log bounding and failure reporting.

Key Points

  • Agents and CLI runtimes now recover cleanly from interrupted tool calls, stale session bindings, compaction handoffs, and media delivery retries with improved session lock management and orphan state cleanup
  • Multi-channel delivery (Telegram, WhatsApp, iMessage, Slack, Discord, Teams, Google Chat/Meet) is more stable with capped request/retry timers and improved WebSocket ping paths for mobile sessions
  • Provider and plugin requests now have bounded timers, retries, OAuth/device-code lifetimes, media downloads, and polling paths to prevent runs from hanging
  • New Skill Workshop feature enables governed skill creation with reviewable proposals, CLI/Gateway review actions, versioned proposal frontmatter, and rollback safeguards
  • Tokenjuice and GitHub Copilot are externalized as official npm plugins (@openclaw/tokenjuice, @openclaw/copilot) with ClawHub publish metadata
  • Workboard adds multi-agent orchestration primitives and coordination tools for planning and run tracking
  • iOS push relay now includes hosted defaults, realtime Talk playback, and guarded WebSocket ping paths for reliable mobile sessions
  • CI/E2E infrastructure improvements cap logs, response bodies, readiness probes, and artifact checks to prevent stalling and provide bounded failure proof
  • Skills runtime loading, status filtering, and prompt formatting are centralized through a core skills index
  • Security hardening rejects unsafe OAuth/token lifetimes, retry-after delays, response body sizes, and command timeout configurations

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

[Release] openclaw/openclaw v2026.5.31-beta.1: openclaw 2026.5.31-beta.1 | Agent Daily