feat: enhance README and UI for team management and review processes

- Added "Flexible autonomy" feature to README, allowing users to configure agent control levels.
- Updated task creation description for clarity in README.
- Improved team detail view messaging to better inform users about draft teams and provisioning status.
- Enhanced review function to prevent duplicate approval comments and events, ensuring idempotency.
- Expanded expected IPC signals to include 'TEAM_DRAFT' for better error handling.

README.md

@@ -39,12 +39,13 @@ A new approach to task management with AI agent teams.
 - **Task-specific logs and messages** — clearly see all Claude logs and messages in isolation for each individual task, making it easy to trace what happened for any assignment
 - **Live process section** — see which agents are running processes and open URLs directly in the browser
 - **Stay in control** — send a direct message to any agent, drop a comment on a task, or pick a quick action right on the kanban card whenever you want to clarify something or add new work
+- **Flexible autonomy** — let agents run fully autonomous, or review and approve each action one by one (you'll get a notification) — configure the level of control that fits your security needs
 - **Solo mode** — one-member team: a single agent that creates its own tasks and shows live progress. Saves tokens; can expand to a full team anytime
 
 <details>
 <summary><strong>More features</strong></summary>
 
-- **Task creation with attachments** — Simply send a message to the team lead with any attached images (planed all files). The lead will automatically create a fully described task and attach your files directly to the task for complete context.
+- **Task creation with attachments** — send a message to the team lead with any attached images. The lead will automatically create a fully described task and attach your files directly to the task for complete context.
 
 - **Deep session analysis** — detailed breakdown of what happened in each Claude session: bash commands, reasoning, subprocesses
 
@@ -64,7 +65,7 @@ A new approach to task management with AI agent teams.
 
 - **Attach code context** — reference files or snippets in messages, like in Cursor. You can also mention tasks using `#task-id`, or refer to another team with `@team-name` in your messages.
 
-- **Notification system** — configurable alerts when tasks complete, agents need attention, or errors occur
+- **Notification system** — configurable alerts when tasks complete, agents need your response, new comments arrive, or errors occur
 
 - **MCP integration** — supports the built-in `mcp-server` (see [mcp-server folder](./mcp-server)) for integrating external tools and extensible agent plugins out of the box
 
@@ -117,7 +118,7 @@ No prerequisites — Claude Code can be installed and configured directly from t
 
 ### Tech stack
 
-Electron, React 18, TypeScript 5, Tailwind CSS 3, Zustand 4. Data from `~/.claude/` (session logs, todos, tasks). No cloud backend — everything runs locally.
+Electron 40, React 18, TypeScript 5, Tailwind CSS 3, Zustand 4. Data from `~/.claude/` (session logs, todos, tasks). No cloud backend — everything runs locally.
 
 ---
 

agent-teams-controller/src/internal/review.js

@@ -153,6 +153,11 @@ function approveReview(context, taskId, flags = {}) {
   const leadSessionId = resolveLeadSessionId(context, flags);
   const prevReviewState = getCurrentReviewState(task);
 
+  // Idempotent: already approved → skip duplicate comment/event, only add note if new
+  if (prevReviewState === 'approved') {
+    return { ok: true, taskId: task.id, displayId: task.displayId, column: 'approved', alreadyApproved: true };
+  }
+
   kanban.setKanbanColumn(context, task.id, 'approved');
 
   // Append review_approved event

src/renderer/components/team/TeamDetailView.tsx

@@ -971,8 +971,10 @@ export const TeamDetailView = ({ teamName }: TeamDetailViewProps): React.JSX.Ele
           <div className="max-w-md text-center">
             <p className="text-sm font-medium text-text">Team not launched yet</p>
             <p className="mt-2 text-xs text-text-secondary">
-              {teamSummary?.displayName || teamName} configuration has been saved. Launch to start
-              provisioning with CLI.
+              This is a draft team — <strong>{teamSummary?.displayName || teamName}</strong> has
+              been configured with {teamSummary?.memberCount ?? 0} member
+              {teamSummary?.memberCount === 1 ? '' : 's'} but hasn&apos;t been provisioned by CLI
+              yet. Click Launch to select a model and start the team.
             </p>
             <div className="mt-4 flex justify-center gap-2">
               <button
@@ -1022,8 +1024,13 @@ export const TeamDetailView = ({ teamName }: TeamDetailViewProps): React.JSX.Ele
 
   if (!data) {
     return (
-      <div className="flex size-full items-center justify-center p-6 text-sm text-[var(--color-text-muted)]">
-        No team data available
+      <div className="size-full overflow-auto p-4">
+        <div ref={provisioningBannerRef}>
+          <TeamProvisioningBanner teamName={teamName} />
+        </div>
+        <div className="flex flex-1 items-center justify-center p-6 text-sm text-[var(--color-text-muted)]">
+          No team data available
+        </div>
       </div>
     );
   }

src/renderer/utils/unwrapIpc.ts

@@ -14,7 +14,7 @@ export class IpcError extends Error {
 }
 
 /** Error messages that represent expected transient states, not real failures. */
-const EXPECTED_IPC_SIGNALS = ['TEAM_PROVISIONING'];
+const EXPECTED_IPC_SIGNALS = ['TEAM_PROVISIONING', 'TEAM_DRAFT'];
 
 export async function unwrapIpc<T>(operation: string, fn: () => Promise<T>): Promise<T> {
   try {