diff --git a/packages/app/src/components/core/timeline/TimelineToolbar.tsx b/packages/app/src/components/core/timeline/TimelineToolbar.tsx
new file mode 100644
index 00000000..0c4f6f37
--- /dev/null
+++ b/packages/app/src/components/core/timeline/TimelineToolbar.tsx
@@ -0,0 +1,124 @@
+'use client'
+
+import { useState } from 'react'
+import { useTimeline } from '@aitube/timeline'
+import { ClapSegmentCategory } from '@aitube/clap'
+import { Plus, ListPlus } from 'lucide-react'
+
+import { cn } from '@/lib/utils'
+import {
+  Select,
+  SelectContent,
+  SelectItem,
+  SelectTrigger,
+  SelectValue,
+} from '@/components/ui/select'
+
+const TRACK_CATEGORIES: { value: ClapSegmentCategory; label: string }[] = [
+  { value: ClapSegmentCategory.VIDEO, label: 'Video' },
+  { value: ClapSegmentCategory.IMAGE, label: 'Image' },
+  { value: ClapSegmentCategory.DIALOGUE, label: 'Dialogue' },
+  { value: ClapSegmentCategory.MUSIC, label: 'Music' },
+  { value: ClapSegmentCategory.SOUND, label: 'Sound' },
+  { value: ClapSegmentCategory.CAMERA, label: 'Camera' },
+  { value: ClapSegmentCategory.ACTION, label: 'Action' },
+  { value: ClapSegmentCategory.CHARACTER, label: 'Character' },
+  { value: ClapSegmentCategory.LOCATION, label: 'Location' },
+  { value: ClapSegmentCategory.LIGHTING, label: 'Lighting' },
+  { value: ClapSegmentCategory.STYLE, label: 'Style' },
+]
+
+export function TimelineToolbar() {
+  const [selectedCategory, setSelectedCategory] = useState<ClapSegmentCategory>(
+    ClapSegmentCategory.VIDEO
+  )
+  const [selectedTrack, setSelectedTrack] = useState<string>('')
+
+  const tracks = useTimeline((s) => s.tracks)
+  const createTrack = useTimeline((s) => s.createTrack)
+  const createClip = useTimeline((s) => s.createClip)
+  const cursorTimestampAtInMs = useTimeline((s) => s.cursorTimestampAtInMs)
+
+  const handleCreateTrack = () => {
+    const newTrackId = createTrack(selectedCategory)
+    setSelectedTrack(String(newTrackId))
+  }
+
+  const handleCreateClip = async () => {
+    const trackNumber = selectedTrack ? parseInt(selectedTrack, 10) : -1
+    if (trackNumber < 0 || !tracks[trackNumber]) return
+
+    await createClip({
+      track: trackNumber,
+      startTimeInMs: cursorTimestampAtInMs || 0,
+      category: selectedCategory,
+    })
+  }
+
+  return (
+    <div
+      className={cn(
+        'flex flex-row items-center gap-2',
+        'h-7 px-2',
+        'text-xs text-white/70'
+      )}
+    >
+      <Select
+        value={selectedCategory}
+        onValueChange={(val) => setSelectedCategory(val as ClapSegmentCategory)}
+      >
+        <SelectTrigger className="h-5 w-[110px] border-white/20 bg-transparent text-xs">
+          <SelectValue placeholder="Category" />
+        </SelectTrigger>
+        <SelectContent>
+          {TRACK_CATEGORIES.map(({ value, label }) => (
+            <SelectItem key={value} value={value}>
+              {label}
+            </SelectItem>
+          ))}
+        </SelectContent>
+      </Select>
+
+      <button
+        onClick={handleCreateTrack}
+        className={cn(
+          'flex items-center gap-1 rounded px-1.5 py-0.5',
+          'bg-white/10 hover:bg-white/20',
+          'transition-colors'
+        )}
+        title="Create a new track with the selected category"
+      >
+        <ListPlus className="h-3.5 w-3.5" />
+        <span>Track</span>
+      </button>
+
+      <Select value={selectedTrack} onValueChange={setSelectedTrack}>
+        <SelectTrigger className="h-5 w-[130px] border-white/20 bg-transparent text-xs">
+          <SelectValue placeholder="Select track" />
+        </SelectTrigger>
+        <SelectContent>
+          {tracks.map((track) => (
+            <SelectItem key={track.id} value={String(track.id)}>
+              Track {track.id} ({track.name})
+            </SelectItem>
+          ))}
+        </SelectContent>
+      </Select>
+
+      <button
+        onClick={handleCreateClip}
+        disabled={!selectedTrack}
+        className={cn(
+          'flex items-center gap-1 rounded px-1.5 py-0.5',
+          'bg-white/10 hover:bg-white/20',
+          'transition-colors',
+          'disabled:cursor-not-allowed disabled:opacity-40'
+        )}
+        title="Create a new clip on the selected track at the cursor position"
+      >
+        <Plus className="h-3.5 w-3.5" />
+        <span>Clip</span>
+      </button>
+    </div>
+  )
+}
diff --git a/packages/app/src/components/core/timeline/index.tsx b/packages/app/src/components/core/timeline/index.tsx
index 2a15af9c..719b8bae 100644
--- a/packages/app/src/components/core/timeline/index.tsx
+++ b/packages/app/src/components/core/timeline/index.tsx
@@ -4,6 +4,8 @@ import { ClapTimeline, useTimeline, SegmentResolver } from '@aitube/timeline'
 import { useMonitor } from '@/services/monitor/useMonitor'
 import { useResolver } from '@/services/resolver/useResolver'
 import { useUI } from '@/services/ui'
+import { cn } from '@/lib/utils'
+import { TimelineToolbar } from './TimelineToolbar'
 
 export function Timeline(
   {
@@ -56,13 +58,12 @@ export function Timeline(
     togglePlayback,
   ])
 
-  if (className) {
-    return (
-      <div className={className}>
+  return (
+    <div className={cn('flex h-full w-full flex-col', className)}>
+      <TimelineToolbar />
+      <div className="flex-1 overflow-hidden">
         <ClapTimeline showFPS={false} />
       </div>
-    )
-  }
-
-  return <ClapTimeline showFPS={false} className={className} />
+    </div>
+  )
 }
diff --git a/packages/timeline/src/hooks/useTimeline.ts b/packages/timeline/src/hooks/useTimeline.ts
index c5276f4a..c467bf74 100644
--- a/packages/timeline/src/hooks/useTimeline.ts
+++ b/packages/timeline/src/hooks/useTimeline.ts
@@ -1,7 +1,7 @@
 import { create } from "zustand"
 import * as THREE from "three"
 import type { ThreeEvent } from "@react-three/fiber"
-import { ClapProject, ClapSegment, ClapSegmentCategory, isValidNumber, newClap, serializeClap, ClapTracks, ClapEntity, ClapMeta } from "@aitube/clap"
+import { ClapProject, ClapSegment, ClapSegmentCategory, ClapOutputType, isValidNumber, newClap, newSegment, serializeClap, ClapTracks, ClapEntity, ClapMeta } from "@aitube/clap"
 
 import { TimelineSegment, SegmentEditionStatus, SegmentVisibility, TimelineStore, SegmentArea, SegmentPointerEvent, SegmentEventCallbackHandler, Invalidate } from "@/types/timeline"
 import { getDefaultProjectState, getDefaultState } from "@/utils/getDefaultState"
@@ -1206,6 +1206,198 @@ export const useTimeline = create<TimelineStore>((set, get) => ({
       })
     })
   },
+  createTrack: (category: ClapSegmentCategory): number => {
+    const {
+      width,
+      height,
+      tracks,
+      cellWidth,
+      defaultSegmentDurationInSteps,
+      durationInMsPerStep,
+      durationInMs,
+      defaultPreviewHeight,
+      defaultCellHeight,
+    } = get()
+
+    const isPreview =
+      category === ClapSegmentCategory.IMAGE ||
+      category === ClapSegmentCategory.VIDEO
+
+    const newTrackId = tracks.length
+
+    const updatedTracks = [
+      ...tracks,
+      {
+        id: newTrackId,
+        name: `${category}`,
+        isPreview,
+        height: isPreview ? defaultPreviewHeight : defaultCellHeight,
+        hue: 0,
+        occupied: true,
+        visible: true,
+      },
+    ]
+
+    set({
+      ...computeContentSizeMetrics({
+        width,
+        height,
+        tracks: updatedTracks,
+        cellWidth,
+        defaultSegmentDurationInSteps,
+        durationInMsPerStep,
+        durationInMs,
+      }),
+    })
+
+    return newTrackId
+  },
+
+  createClip: async ({
+    track,
+    startTimeInMs,
+    endTimeInMs,
+    category,
+    prompt = '',
+  }: {
+    track: number
+    startTimeInMs: number
+    endTimeInMs?: number
+    category?: ClapSegmentCategory
+    prompt?: string
+  }): Promise<TimelineSegment> => {
+    const {
+      tracks,
+      durationInMsPerStep,
+      defaultSegmentDurationInSteps,
+      addSegment,
+    } = get()
+
+    const trackInfo = tracks[track]
+    const trackCategory = category || (
+      trackInfo
+        ? ClapSegmentCategory[trackInfo.name as keyof typeof ClapSegmentCategory] || ClapSegmentCategory.GENERIC
+        : ClapSegmentCategory.GENERIC
+    )
+
+    const defaultDurationInMs = defaultSegmentDurationInSteps * durationInMsPerStep
+    const segmentEndTimeInMs = endTimeInMs || (startTimeInMs + defaultDurationInMs)
+
+    const outputType =
+      trackCategory === ClapSegmentCategory.VIDEO ? ClapOutputType.VIDEO
+      : trackCategory === ClapSegmentCategory.IMAGE ? ClapOutputType.IMAGE
+      : trackCategory === ClapSegmentCategory.DIALOGUE ? ClapOutputType.AUDIO
+      : trackCategory === ClapSegmentCategory.MUSIC ? ClapOutputType.AUDIO
+      : trackCategory === ClapSegmentCategory.SOUND ? ClapOutputType.AUDIO
+      : ClapOutputType.TEXT
+
+    // Check for time-overlap collisions on the requested track
+    const { segments, findFreeTrack: findFree } = get()
+    const hasCollision = segments.some(
+      (s) =>
+        s.track === track &&
+        !(s.endTimeInMs <= startTimeInMs || s.startTimeInMs >= segmentEndTimeInMs)
+    )
+
+    const resolvedTrack = hasCollision
+      ? findFree({ startTimeInMs, endTimeInMs: segmentEndTimeInMs })
+      : track
+
+    const clapSegment = newSegment({
+      track: resolvedTrack,
+      startTimeInMs,
+      endTimeInMs: segmentEndTimeInMs,
+      category: trackCategory,
+      prompt,
+      outputType,
+    })
+
+    const segment = await clapSegmentToTimelineSegment(clapSegment)
+
+    await addSegment({ segment, startTimeInMs, track: resolvedTrack })
+
+    return segment
+  },
+
+  moveSegmentToTrack: (segment: TimelineSegment, targetTrack: number): boolean => {
+    const {
+      width,
+      height,
+      tracks,
+      segments,
+      cellWidth,
+      defaultSegmentDurationInSteps,
+      durationInMsPerStep,
+      durationInMs,
+      invalidate,
+      allSegmentsChanged: prevAllChanged,
+      atLeastOneSegmentChanged: prevOneChanged,
+    } = get()
+
+    const targetTrackInfo = tracks[targetTrack]
+    if (!targetTrackInfo) { return false }
+
+    // Allow moves to empty/misc tracks or same-category tracks
+    const targetName = targetTrackInfo.name
+    const isCompatible =
+      !targetTrackInfo.occupied ||
+      targetName === '(empty)' ||
+      targetName === '(misc)' ||
+      targetName === segment.category
+
+    if (!isCompatible) { return false }
+
+    // Check for time-overlap collisions on the target track
+    const hasCollision = segments.some(
+      (s) =>
+        s.id !== segment.id &&
+        s.track === targetTrack &&
+        !(s.endTimeInMs <= segment.startTimeInMs || s.startTimeInMs >= segment.endTimeInMs)
+    )
+
+    if (hasCollision) { return false }
+
+    // Immutable update: create new segment with updated track
+    const updatedSegments = segments.map((s) =>
+      s.id === segment.id ? { ...s, track: targetTrack } : s
+    )
+
+    // Immutable update: create new tracks array if target was empty
+    const updatedTracks = !targetTrackInfo.occupied
+      ? tracks.map((t, i) => {
+          if (i !== targetTrack) { return t }
+          const isPreview =
+            segment.category === ClapSegmentCategory.IMAGE ||
+            segment.category === ClapSegmentCategory.VIDEO
+          return {
+            ...t,
+            name: `${segment.category}`,
+            occupied: true,
+            isPreview,
+          }
+        })
+      : [...tracks]
+
+    set({
+      segments: updatedSegments,
+      allSegmentsChanged: prevAllChanged + 1,
+      atLeastOneSegmentChanged: prevOneChanged + 1,
+      ...computeContentSizeMetrics({
+        width,
+        height,
+        tracks: updatedTracks,
+        cellWidth,
+        defaultSegmentDurationInSteps,
+        durationInMsPerStep,
+        durationInMs,
+      }),
+    })
+
+    invalidate()
+
+    return true
+  },
+
   deleteSegments: (ids: string[]): void => {
     const {
       segments: previousSegments,
diff --git a/packages/timeline/src/types/timeline.ts b/packages/timeline/src/types/timeline.ts
index 71f9a4f7..e6d1e057 100644
--- a/packages/timeline/src/types/timeline.ts
+++ b/packages/timeline/src/types/timeline.ts
@@ -1,6 +1,6 @@
 import * as THREE from "three"
 import type { ThreeEvent } from "@react-three/fiber"
-import { ClapEntity, ClapImageRatio, ClapMeta, ClapProject, ClapScene, ClapSegment, ClapTracks } from "@aitube/clap"
+import { ClapEntity, ClapImageRatio, ClapMeta, ClapProject, ClapScene, ClapSegment, ClapSegmentCategory, ClapTracks } from "@aitube/clap"
 
 import { ClapSegmentColorScheme, ClapTimelineTheme } from "./theme"
 import { TimelineControlsImpl } from "@/components/controls/types"
@@ -397,6 +397,29 @@ export type TimelineStoreModifiers = {
    */
   deleteSegments: (ids: string[]) => void
 
+  /**
+   * Create a new track with a given category
+   * @returns the track number of the newly created track
+   */
+  createTrack: (category: ClapSegmentCategory) => number
+
+  /**
+   * Create a new clip/segment on a track at a given time position
+   */
+  createClip: (params: {
+    track: number
+    startTimeInMs: number
+    endTimeInMs?: number
+    category?: ClapSegmentCategory
+    prompt?: string
+  }) => Promise<TimelineSegment>
+
+  /**
+   * Move a segment to a different track, validating category compatibility
+   * @returns true if the move succeeded
+   */
+  moveSegmentToTrack: (segment: TimelineSegment, targetTrack: number) => boolean
+
   addEntities: (entities: ClapEntity[]) => Promise<void>
   updateEntities: (entities: ClapEntity[]) => Promise<void>
   deleteEntities: (entities: (ClapEntity|string)[]) => Promise<void>