Open the Web Shell sidebar activity view and Add to Split View
Qwen Code v0.26.6 (no known breaking changes) adds a trusted-workspace sidebar activity view, Add to Split View, Token Usage success/speed/latency by model, a dismissible Workflow chip in the composer + menu, channel /approve|/deny for one-time tools, and headless prompt.submit hooks.Why this loop
v0.26.6 has no known breaking changes and is mainly a Web Shell operations upgrade. Trusted-workspace conversations gain a sidebar activity view that shows approval, input, and error states, unread completions, and running work, with bulk read/archive, so the team can triage agent work without opening every session. Eligible sidebar session menus add Add to Split View for side-by-side sessions. Token Usage now reports request success rate, output speed, per-model success indicators and distinct colors, and plots per-request output speed, first-token latency, and failure details by model, so default-model choices can follow session evidence. Workflow mode moved from the composer toolbar into the + menu; a dismissible chip turns it off and warns when tools are unavailable. Channel users can /approve or /deny their own pending one-time tool permission requests in the current attended turn. Headless -p and stream-json runs can fire mod prompt.submit hooks to inspect or rewrite prompts before submission. MCP discovery keeps uniquely named tools after a name collision instead of hiding an entire server.
Proposed actions
- Upgrade Qwen Code to v0.26.6 from https://github.com/QwenLM/qwen-code/releases/tag/v0.26.6; on Fedora 44 or RHEL 10 x64 use the newly published desktop RPM.
- In Web Shell, open a trusted-workspace conversation, pin the sidebar activity view, triage approval/input/error states first, then bulk-read unread completions and bulk-archive finished running work.
- On an eligible sidebar session menu, choose Add to Split View and keep two related sessions (for example implement vs review) visible side by side instead of tab-switching.
- Open the Web Shell Token Usage panel and compare request success rate, per-request output speed, first-token latency, per-model success indicators, distinct model colors, and failure details; pin the session default to the model with the better speed and success on your workload.
- In GitHub/channel attended turns, resolve your own pending one-time tool permission requests with /approve or /deny in the current turn instead of waiting on another operator.
Agent prompt
Paste into your agent or query via MCP (get_agent_prompt) — free, no extra AI costCopilot Chat / Agent mode
DevAgentRadar → GitHub Copilot
You are helping me adopt a real coding-assistant change. Work only from the facts below. Do not invent features.
Context
Assistant: Qwen Code Proposal: Open the Web Shell sidebar activity view and Add to Split View Summary: Qwen Code v0.26.6 (no known breaking changes) adds a trusted-workspace sidebar activity view, Add to Split View, Token Usage success/speed/latency by model, a dismissible Workflow chip in the composer + menu, channel /approve|/deny for one-time tools, and headless prompt.submit hooks. Primary source: https://github.com/QwenLM/qwen-code/releases/tag/v0.26.6
Why it matters
v0.26.6 has no known breaking changes and is mainly a Web Shell operations upgrade. Trusted-workspace conversations gain a sidebar activity view that shows approval, input, and error states, unread completions, and running work, with bulk read/archive, so the team can triage agent work without opening every session. Eligible sidebar session menus add Add to Split View for side-by-side sessions. Token Usage now reports request success rate, output speed, per-model success indicators and distinct colors, and plots per-request output speed, first-token latency, and failure details by model, so default-model choices can follow session evidence. Workflow mode moved from the composer toolbar into the + menu; a dismissible chip turns it off and warns when tools are unavailable. Channel users can /approve or /deny their own pending one-time tool permission requests in the current attended turn. Headless -p and stream-json runs can fire mod prompt.submit hooks to inspect or rewrite prompts before submission. MCP discovery keeps uniquely named tools after a name collision instead of hiding an entire server.
Suggested actions
- Upgrade Qwen Code to v0.26.6 from https://github.com/QwenLM/qwen-code/releases/tag/v0.26.6; on Fedora 44 or RHEL 10 x64 use the newly published desktop RPM.
- In Web Shell, open a trusted-workspace conversation, pin the sidebar activity view, triage approval/input/error states first, then bulk-read unread completions and bulk-archive finished running work.
- On an eligible sidebar session menu, choose Add to Split View and keep two related sessions (for example implement vs review) visible side by side instead of tab-switching.
- Open the Web Shell Token Usage panel and compare request success rate, per-request output speed, first-token latency, per-model success indicators, distinct model colors, and failure details; pin the session default to the model with the better speed and success on your workload.
- In GitHub/channel attended turns, resolve your own pending one-time tool permission requests with /approve or /deny in the current turn instead of waiting on another operator.
After you finish
Do not report this as applied to DevAgentRadar. You cannot write the visitor's loop.
Tell the human: open https://devagentradar.com/proposals/qwen-code-v0-26-6-open-the-web-shell-sidebar-activity-view-and-add-to-sp and mark Applied, Skipped, or Failed. Proposal id: c2417ca8-91e5-4ce0-8e8d-1af7ddd9bdf8
Your job
- Restate the change in one sentence.
- Propose a minimal plan for my repo (or a throwaway pilot).
- Implement only what I approve; prefer small diffs and tests.
- Call out risks (permissions, breaking APIs, cost).
Start by confirming you understood the proposal.
Your loop
This browser · no sign-in · not shared as “you”After you run the prompt
Only you can mark this. Agents cannot write your loop.
Your decision stays on this device. A public tally appears after a few votes.