Adopt multi-session visibility and free-tier models in Cline Desktop
Use the system tray, favorites, paginated history, and subagent run views to manage concurrent agent work, and try free Cline models with clear limit messaging before paid usage.Why this loop
Cline Desktop v0.0.7 makes concurrent agent work observable: a system tray icon reports app status and how many sessions are running, history loads in 10-session pages (older pages only when you hit the end), favorites plus most-recent-activity ordering keep live work findable, status-dot colors are consistent across views, and subagent/teammate runs from a session now appear with status and results. Free Cline models are labeled (free) in pickers and show a reset time when you hit the free-tier limit, so you can trial before paid spend. Regional China/international toggles for Qwen, Moonshot, Z AI, and MiniMax were previously ignored and silently sent traffic to the wrong host—re-verify after upgrade. Tool calls no longer fail when a model emits a line number as a string (e.g. insert_line: "3"), which used to force a wasted retry. Chat polish (tool-specific disclosure icons, elapsed thinking time, aligned timestamps, non-shifting message actions, viewport-anchored scroll) plus a refreshed provider/model catalog reduce noise while you run more sessions.
Proposed actions
- Install or update Cline Desktop to desktop-v0.0.7 from https://github.com/cline/cline/releases/tag/desktop-v0.0.7. After launch, confirm the system tray icon is visible; start two agent sessions and check that the tray reports app status and a running-session count of 2.
- In the Cline Desktop model picker, choose a Cline model labeled (free), run a short task, and if you hit the free-tier cap confirm the UI shows a reset time (not a generic error) before switching to a paid model.
- Open session history, favorite at least one active session, scroll to the end of the first 10-session page so the next page fetches, and confirm sessions are ordered by most recent activity with matching status-dot colors across views.
- From a parent session that spawned subagent or teammate runs, open that session and confirm each nested run appears in the app with its status and results.
- If you use Qwen, Moonshot, Z AI, or MiniMax with China vs international endpoints, re-apply the regional toggle on desktop-v0.0.7 and send a test request to confirm traffic hits the intended host (v0.0.6 ignored the toggle).
Agent prompt
Paste into your agent or query via MCP (get_agent_prompt) — free, no extra AI costCline / .clinerules task
DevAgentRadar → Cline
You are helping me adopt a real coding-assistant change. Work only from the facts below. Do not invent features.
Context
Assistant: Cline Proposal: Adopt multi-session visibility and free-tier models in Cline Desktop Summary: Use the system tray, favorites, paginated history, and subagent run views to manage concurrent agent work, and try free Cline models with clear limit messaging before paid usage. Primary source: https://github.com/cline/cline/releases/tag/desktop-v0.0.7
Why it matters
Cline Desktop v0.0.7 makes concurrent agent work observable: a system tray icon reports app status and how many sessions are running, history loads in 10-session pages (older pages only when you hit the end), favorites plus most-recent-activity ordering keep live work findable, status-dot colors are consistent across views, and subagent/teammate runs from a session now appear with status and results. Free Cline models are labeled (free) in pickers and show a reset time when you hit the free-tier limit, so you can trial before paid spend. Regional China/international toggles for Qwen, Moonshot, Z AI, and MiniMax were previously ignored and silently sent traffic to the wrong host—re-verify after upgrade. Tool calls no longer fail when a model emits a line number as a string (e.g. insert_line: "3"), which used to force a wasted retry. Chat polish (tool-specific disclosure icons, elapsed thinking time, aligned timestamps, non-shifting message actions, viewport-anchored scroll) plus a refreshed provider/model catalog reduce noise while you run more sessions.
Suggested actions
- Install or update Cline Desktop to desktop-v0.0.7 from https://github.com/cline/cline/releases/tag/desktop-v0.0.7. After launch, confirm the system tray icon is visible; start two agent sessions and check that the tray reports app status and a running-session count of 2.
- In the Cline Desktop model picker, choose a Cline model labeled (free), run a short task, and if you hit the free-tier cap confirm the UI shows a reset time (not a generic error) before switching to a paid model.
- Open session history, favorite at least one active session, scroll to the end of the first 10-session page so the next page fetches, and confirm sessions are ordered by most recent activity with matching status-dot colors across views.
- From a parent session that spawned subagent or teammate runs, open that session and confirm each nested run appears in the app with its status and results.
- If you use Qwen, Moonshot, Z AI, or MiniMax with China vs international endpoints, re-apply the regional toggle on desktop-v0.0.7 and send a test request to confirm traffic hits the intended host (v0.0.6 ignored the toggle).
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/cline-desktop-v0-0-7-adopt-multi-session-visibility-and-free-tier-models and mark Applied, Skipped, or Failed. Proposal id: cbd72a44-7442-4fc0-86eb-47962d6a958c
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.