OpenClaw: Plugin Subagents, Profile Isolation, and Platform Cleanup

Plugins gain native subagent support and requester-scoped model catalogs, profile and auth reliability issues get fixed, and a broad wave of refactors and test consolidation trims maintenance cost.

Duration: PT2M35S

Episode overview

This episode is a short developer briefing from OpenClaw.

It explains recent repository work in plain language.

  • Show: OpenClaw
  • Published: 2026-10-22T06:06:09Z
  • Audio duration: PT2M35S

Transcript excerpt

This excerpt keeps the crawler page concise. Listen to the episode or use the RSS feed for the full update.

Good morning, it's October 22nd, 2026, and this is OpenClaw.

The throughline today is plugins growing up, and the platform getting steadier underneath them.

First theme: plugin autonomy. PR 180173 lets authenticated plugin commands launch native helper sessions under the current requester's authority, with command text preserved and stop handling intact. PR 180274 closes the loop, adding a list call so a plugin that crashes after a spawn can recover children it launched…

Second: reliability fixes for multi-profile and production load. PR 180288 gives each named profile its own node service across Mac, Linux and Windows, so installs and force reinstalls no longer collide with the default. PR 180280 fixes auth cooldowns that could lock a profile out for up to a day after routine rate…

Third: efficiency and cleanup. PR 180149 moves workspace checkpoints to incremental manifests with a shared-state schema bump to 31, cutting repeated full transfers. PR 180142 trims allocation work in session broadcasts, and related startup work moves credential preparation off the main thread. Meanwhile a wave of…

What's next: if you run named profiles, verify node services after update, and if you…

Nearby episodes from OpenClaw

  1. Worker Reclaim, Restart Recovery, and UI Stability
  2. Gateway Startup and Session Recovery Fixes
  3. Deep Cleanup, UI Polish, and Router Expansion
  4. PostgreSQL Goes Supported Amid Major Cleanup
  5. Gateway Resilience and CLI Backend Stability
  6. Honest Failures, Smarter Tool Checks, and Memory Cuts
  7. Restart Recovery Hardening
  8. Solid UI Cleanup and Repair Reliability