[Release] openclaw/openclaw v2026.5.7: openclaw 2026.5.7

OpenClaw v2026.5.7 is a maintenance release focused on stability, security, and reliability improvements across plugin publishing, CLI tools, authentication, and multi-channel integrations. Key fixes address cron job persistence, Discord/Telegram/WhatsApp message routing, OAuth recovery, permission enforcement, and context caching issues. The release includes enhancements to CLI output formatting, voice capture quality, and error handling to prevent partial publishes and stale session states.

Key Points

• •Plugin publishing now retries transient dependency failures and verifies all ClawHub package versions post-publish to prevent partial publishes and speed maintenance recovery
• •Cron CLI enhanced with computed status output (disabled/running/ok/error/skipped/idle) in JSON mode for external tooling integration without reimplementing status logic
• •Channels CLI refactored to separate channel-only listing from bundled/catalog channels via `--all` flag, with auth details moved to dedicated `models auth list` and `status` commands
• •Security improvements: native command owner enforcement, admin-only global memory toggles, and authorization hooks for inline skill tool dispatch
• •Gateway sessions now clear cached skill snapshots on `/new` and `reset` to rebuild visible skill lists after changes, and persist new transcript files on daily rollover
• •Discord/Telegram/WhatsApp routing fixes: provider-prefixed targets parsed correctly, polling watchdog tied to `getUpdates` liveness, and LID forward mappings applied for WhatsApp sends
• •Context engine invalidates cached assembled views when source history shrinks or assembly fails, preventing stale pre-reset history reuse
• •Cron job repair via `doctor --fix` handles malformed `payload.model` values (null, blank, 'default') while maintaining strict runtime validation
• •Compaction summary reserve tokens clamped to model output limits, preventing invalid `max_tokens` requests in high-context scenarios
• •Voice capture improved with extended 2.5s post-speech silence grace period and configurable `voice.captureSilenceGraceMs` for noisy environments

Workflow Diagram

Concepts