diff --git a/packages/app/src/composer/agent-controls/mode-control.tsx b/packages/app/src/composer/agent-controls/mode-control.tsx
index 6329cd46e..e129b1e53 100644
--- a/packages/app/src/composer/agent-controls/mode-control.tsx
+++ b/packages/app/src/composer/agent-controls/mode-control.tsx
@@ -12,7 +12,16 @@ import { Text, View, type PressableStateCallbackType } from "react-native";
 import { StyleSheet, useUnistyles } from "react-native-unistyles";
 import { useShallow } from "zustand/shallow";
 import { useStoreWithEqualityFn } from "zustand/traditional";
-import { Bot, ShieldAlert, ShieldCheck, ShieldOff, ShieldQuestionMark } from "lucide-react-native";
+import {
+  Bot,
+  Shield,
+  ShieldAlert,
+  ShieldCheck,
+  ShieldEllipsis,
+  ShieldOff,
+  ShieldPlus,
+  ShieldQuestionMark,
+} from "lucide-react-native";
 import { ComboboxTrigger } from "@/components/ui/combobox-trigger";
 import { type SheetHeader } from "@/components/adaptive-modal-sheet";
 import { Combobox, ComboboxItem, type ComboboxOption } from "@/components/ui/combobox";
@@ -48,9 +57,12 @@ interface ModeIconProps {
 
 const MODE_ICONS: Record<string, ComponentType<ModeIconProps>> = {
   Bot,
+  Shield,
   ShieldCheck,
   ShieldAlert,
+  ShieldEllipsis,
   ShieldOff,
+  ShieldPlus,
   ShieldQuestionMark,
 };
 
diff --git a/packages/protocol/src/provider-manifest.ts b/packages/protocol/src/provider-manifest.ts
index 312e70455..35af5b1d9 100644
--- a/packages/protocol/src/provider-manifest.ts
+++ b/packages/protocol/src/provider-manifest.ts
@@ -36,39 +36,39 @@ export interface AgentProviderDefinition {
 }
 
 const CLAUDE_MODES: AgentProviderModeDefinition[] = [
+  {
+    id: "plan",
+    label: "Plan Mode",
+    description: "Analyze the codebase without executing tools or edits",
+    icon: "ShieldEllipsis",
+    colorTier: "planning",
+  },
   {
     id: "default",
     label: "Always Ask",
     description: "Prompts for permission the first time a tool is used",
-    icon: "ShieldCheck",
+    icon: "Shield",
     colorTier: "safe",
   },
-  {
-    id: "auto",
-    label: "Auto mode",
-    description: "Uses a model classifier to review permission prompts automatically",
-    icon: "ShieldQuestionMark",
-    colorTier: "moderate",
-  },
   {
     id: "acceptEdits",
     label: "Accept File Edits",
     description: "Automatically approves edit-focused tools without prompting",
-    icon: "ShieldAlert",
+    icon: "ShieldPlus",
     colorTier: "moderate",
   },
   {
-    id: "plan",
-    label: "Plan Mode",
-    description: "Analyze the codebase without executing tools or edits",
+    id: "auto",
+    label: "Auto mode",
+    description: "Uses a model classifier to review permission prompts automatically",
     icon: "ShieldCheck",
-    colorTier: "planning",
+    colorTier: "moderate",
   },
   {
     id: "bypassPermissions",
     label: "Bypass",
     description: "Skip all permission prompts (use with caution)",
-    icon: "ShieldAlert",
+    icon: "ShieldOff",
     colorTier: "dangerous",
     isUnattended: true,
   },
@@ -79,7 +79,7 @@ const CODEX_MODES: AgentProviderModeDefinition[] = [
     id: "auto",
     label: "Default Permissions",
     description: "Edit files and run commands with Codex's default approval flow.",
-    icon: "ShieldAlert",
+    icon: "Shield",
     colorTier: "moderate",
   },
   {
@@ -87,14 +87,14 @@ const CODEX_MODES: AgentProviderModeDefinition[] = [
     label: "Auto-review",
     description:
       "Same workspace-write permissions as Default, but eligible `on-request` approvals are routed through the auto-reviewer subagent.",
-    icon: "ShieldQuestionMark",
+    icon: "ShieldCheck",
     colorTier: "moderate",
   },
   {
     id: "full-access",
     label: "Full Access",
     description: "Edit files, run commands, and access the network without additional prompts.",
-    icon: "ShieldAlert",
+    icon: "ShieldOff",
     colorTier: "dangerous",
     isUnattended: true,
   },
@@ -105,14 +105,14 @@ const COPILOT_MODES: AgentProviderModeDefinition[] = [
     id: "https://agentclientprotocol.com/protocol/session-modes#agent",
     label: "Agent",
     description: "Default agent mode for conversational interactions",
-    icon: "ShieldAlert",
+    icon: "Shield",
     colorTier: "moderate",
   },
   {
     id: "https://agentclientprotocol.com/protocol/session-modes#plan",
     label: "Plan",
     description: "Plan mode for creating and executing multi-step plans",
-    icon: "ShieldCheck",
+    icon: "ShieldEllipsis",
     colorTier: "planning",
   },
   {
@@ -130,14 +130,14 @@ const OPENCODE_MODES: AgentProviderModeDefinition[] = [
     id: "build",
     label: "Build",
     description: "Allows edits and tool execution for implementation work",
-    icon: "Bot",
+    icon: "Shield",
     colorTier: "moderate",
   },
   {
     id: "plan",
     label: "Plan",
     description: "Read-only planning mode that avoids file edits",
-    icon: "Bot",
+    icon: "ShieldEllipsis",
     colorTier: "planning",
   },
 ];
diff --git a/packages/server/src/server/agent/providers/claude/agent.ts b/packages/server/src/server/agent/providers/claude/agent.ts
index a2dde63a6..60001add3 100644
--- a/packages/server/src/server/agent/providers/claude/agent.ts
+++ b/packages/server/src/server/agent/providers/claude/agent.ts
@@ -280,25 +280,25 @@ const CLAUDE_CAPABILITIES: AgentCapabilityFlags = {
 };
 
 const DEFAULT_MODES: AgentMode[] = [
+  {
+    id: "plan",
+    label: "Plan Mode",
+    description: "Analyze the codebase without executing tools or edits",
+  },
   {
     id: "default",
     label: "Always Ask",
     description: "Prompts for permission the first time a tool is used",
   },
-  {
-    id: "auto",
-    label: "Auto mode",
-    description: "Uses a model classifier to review permission prompts automatically",
-  },
   {
     id: "acceptEdits",
     label: "Accept File Edits",
     description: "Automatically approves edit-focused tools without prompting",
   },
   {
-    id: "plan",
-    label: "Plan Mode",
-    description: "Analyze the codebase without executing tools or edits",
+    id: "auto",
+    label: "Auto mode",
+    description: "Uses a model classifier to review permission prompts automatically",
   },
   {
     id: "bypassPermissions",