Qwen Code ships v0.28.1 preview with GitHub channel adapter and unified chat surface
qwen-code-review-bot · ghdev · 2026-11-21
QwenLM/qwen-code released v0.28.1-preview.1 with 15 changes:
- Channels: GitHub added as the third managed channel adapter, alongside existing DingTalk webhook support
- Web-shell: unified chat surface contract with capability guards, background-task explanations from composer chips, and generated-UI cards that save view state via the host
- Managed-agent: safe recovery for vanished Harness admissions; full-stack E2E test observing a real scheduled /loop fire
- Serve: reserves listener headroom from SSE and reaps stalled writers by default
- Design docs: proposal for idle session suspension toward a stateless Hosted Harness; 20-run parity requirement dropped from loop rollout
A stability-and-plumbing preview release building toward multi-channel hosted agent infrastructure.
More from coding & agent
- Route agents by task type, not fixed rules: not every request needs a frontier LLM call — rohanpaul_ai · 2026-11-21
- OpenClaw's Skill Workshop 2.0 lets your agent review and improve its own skills — heyneighbor · 2026-11-21
- Sentry CEO: coding agents nearly ate 1TB of disk in a week, he's now more against handoff — zeeg · 2026-11-21
- Perplexity Computer comes to Chromium browsers via extension, working across tabs and tools — perplexity_ai · 2026-11-21
- VibeArmory 0.3.0 adds on-device ML to monitor Mac AI agent activity locally — agf2007 · 2026-11-21
- AI-powered research pipeline cuts customer feedback cycle from 10 days to under 2 — pavandavuluri · 2026-11-21