release: v1.5.0 — 9 new core modules, UXP panel, CLI, NLP commands

New features: repeated take detection, YouTube chapter generation,
footage search index, post-production deliverables (VFX/ADR/music cue),
color match, multicam auto-switching, auto zoom keyframes, loudness
match, NLP command parser. 4 new route blueprints (timeline, search,
deliverables, nlp). Full UXP panel for Premiere Pro 25.6+. 8 new MCP
tools. 8 CLI commands. 84 new tests. 6 new social presets (13 total).

Author: SysAdminDoc

CHANGELOG.md

@@ -1,5 +1,32 @@
 # Changelog
 
+## [1.5.0] - 2026-03-23
+
+### Added
+- **Repeated Take Detection** — Jaccard similarity sliding-window to find and remove fumbled/repeated takes from transcriptions
+- **YouTube Chapter Generation** — LLM-powered (Ollama/OpenAI/Anthropic) topic detection and chapter timestamp generation
+- **Footage Search** — Index media library by spoken content; search by keyword or phrase across all clips
+- **Post-Production Deliverables** — Generate VFX sheets, ADR lists, music cue sheets, and asset lists as CSV from sequence data
+- **Color Match** — YCbCr histogram matching to match color profile of one clip to a reference clip
+- **Multicam Auto-Switching** — Diarization-driven cut generation for podcast multicam editing
+- **Auto Zoom Keyframes** — Face-detected keyframe generation for dynamic push-in zoom effect
+- **Loudness Match** — FFmpeg two-pass LUFS normalization to match loudness across multiple clips
+- **NLP Command Parser** — Natural language command dispatch via keyword matching and LLM
+- **Timeline Write-Back** — Apply cuts, markers, and keyframes directly to the active Premiere Pro sequence via ExtendScript
+- **Beat Markers** — Sync detected beat timestamps as sequence markers in Premiere Pro
+- **Batch Rename** — Rename project panel clips with find/replace pattern support
+- **Smart Bins** — Auto-sort project items into bins by rule (name contains, type, duration)
+- **SRT to Native Captions** — Import SRT files as native Premiere Pro caption tracks
+- **Batch Export from Markers** — Export individual clips defined by sequence markers
+- **UXP Panel** — New parallel panel for Premiere Pro 25.6+ using modern UXP APIs (com.opencut.uxp)
+- **Timeline Tab** — New CEP panel tab with all timeline write-back operations
+- **Search & NLP Tab** — New CEP panel tab with footage search and AI command input
+- **Deliverables Section** — Post-production document generation in the Export tab
+
+### Fixed
+- Improved progress callback support in color match, auto zoom, and loudness match operations
+- LLMConfig now properly instantiated as dataclass object across all routes
+
 ## v1.2.0 (2026-03-15)
 
 ### New Features

CLAUDE.md

@@ -74,3 +74,59 @@ Use the sync script to update all at once:
 ```bash
 python scripts/sync_version.py --set 1.3.0
 ```
+
+## UXP Panel Development
+
+The UXP panel at `extension/com.opencut.uxp/` targets Premiere Pro 25.6+.
+
+### Setup
+1. Install Adobe UXP Developer Tool from the Creative Cloud app
+2. In UXP Developer Tool: Add Plugin → select `extension/com.opencut.uxp/`
+3. Click "Load" to sideload the panel into Premiere Pro
+4. The UXP panel connects to the same Python backend on port 5679
+
+### Key differences from CEP panel
+- Uses modern ES modules (`import`/`export`) — no ES3 restrictions
+- Direct `fetch()` API — no HTTP bridge needed
+- Timeline write-back uses `premierepro` UXP module (Premiere 25.6+)
+- CEP panel handles write-back for older versions
+
+### Manifests
+- CEP manifest: `extension/com.opencut.panel/CSXS/manifest.xml`
+- UXP manifest: `extension/com.opencut.uxp/manifest.json`
+- Both must be kept in sync for version numbers
+
+## New Optional Dependency Groups (v1.5.0+)
+
+Install groups individually as needed:
+
+```bash
+# LLM features (chapters, NLP commands)
+pip install openai>=1.0.0          # for OpenAI provider
+pip install anthropic>=0.20.0      # for Anthropic provider
+# Ollama: install from https://ollama.ai (no pip needed)
+
+# Color matching (OpenCV required)
+pip install opencv-python-headless>=4.8.0
+
+# Auto-zoom face detection (OpenCV required)
+pip install opencv-python-headless>=4.8.0
+
+# Footage search (stdlib only — no pip required)
+
+# Deliverables (stdlib csv only — no pip required)
+
+# Loudness matching (FFmpeg required — external tool)
+# Download from https://ffmpeg.org/download.html
+```
+
+## Version Sync
+
+When releasing a new version, update these files:
+- `opencut/__init__.py` — `__version__`
+- `pyproject.toml` — `version`
+- `extension/com.opencut.panel/CSXS/manifest.xml` — `ExtensionBundleVersion` and `Version`
+- `extension/com.opencut.uxp/manifest.json` — `version`
+- `extension/com.opencut.uxp/main.js` — `VERSION` constant
+- `README.md` — version badge
+- `CHANGELOG.md` — new section

DEVELOPMENT.md

@@ -1,6 +1,6 @@
 # OpenCut
 
-![Version](https://img.shields.io/badge/version-1.3.0-blue)
+![Version](https://img.shields.io/badge/version-1.5.0-blue)
 ![License](https://img.shields.io/badge/license-MIT-green)
 ![Platform](https://img.shields.io/badge/platform-Windows%2010%2F11-0078D4)
 ![Python](https://img.shields.io/badge/Python-3.9+-3776AB?logo=python&logoColor=white)
@@ -47,7 +47,7 @@ Clone the repo and run `Install.bat` as Administrator. It handles FFmpeg check,
 
 ## Features
 
-OpenCut v1.3.0 includes **142 API routes**, **6 panel tabs** with **43 sub-tabs**, and covers every major video editing automation task.
+OpenCut v1.5.0 includes **142 API routes**, **7 panel tabs** with **50+ sub-tabs**, and covers every major video editing automation task — now with ~20 new capabilities including Timeline Integration, AI-powered NLP commands, footage search, post-production deliverables, and a UXP panel for Premiere Pro 25.6+.
 
 ### Cut & Clean
 
@@ -58,6 +58,7 @@ OpenCut v1.3.0 includes **142 API routes**, **6 panel tabs** with **43 sub-tabs*
 | Waveform Preview | Visual waveform with draggable threshold line synced to slider | FFmpeg PCM + Canvas |
 | Trim Tool | Set in/out points to extract a clip portion (stream copy or re-encode) | FFmpeg |
 | Full Pipeline | Combined silence + filler removal in one pass | Multi-stage |
+| Repeated Take Detection | Automatically detects when speakers restart a sentence or fumble a phrase. Removes the earlier attempt using transcript similarity analysis (Jaccard overlap) | WhisperX |
 
 ### Captions & Transcription
 
@@ -72,6 +73,8 @@ OpenCut v1.3.0 includes **142 API routes**, **6 panel tabs** with **43 sub-tabs*
 | Translation | Translate captions to 50+ languages | deep-translator |
 | Karaoke Mode | Word-by-word highlight sync for lyrics/captions | Pillow renderer |
 | Transcript Editor | Edit segments in-panel with undo/redo and search | Built-in |
+| YouTube Chapter Generation | Analyze transcript with local (Ollama) or cloud (OpenAI, Anthropic) LLM to identify topic changes. Outputs ready-to-paste chapter timestamps for YouTube descriptions | Ollama / OpenAI / Anthropic |
+| SRT to Native Captions | Import any SRT file as a native Premiere Pro caption track | ExtendScript |
 
 ### Audio Processing
 
@@ -87,6 +90,8 @@ OpenCut v1.3.0 includes **142 API routes**, **6 panel tabs** with **43 sub-tabs*
 | SFX Generator | Procedural tones, sweeps, impacts, noise | NumPy synthesis |
 | AI Music Generation | Text-to-music from prompts | MusicGen (AudioCraft) |
 | Audio Preview Player | Floating player to preview generated audio before importing | Built-in |
+| Loudness Match | FFmpeg two-pass LUFS normalization across multiple clips. Ensures consistent audio levels throughout a sequence | FFmpeg loudnorm |
+| Beat Markers | Export detected beat timestamps as Premiere Pro sequence markers for manual snap-to-beat editing | librosa + ExtendScript |
 
 ### Video Effects & Processing
 
@@ -111,6 +116,7 @@ OpenCut v1.3.0 includes **142 API routes**, **6 panel tabs** with **43 sub-tabs*
 | Merge / Concatenate | Join multiple clips (fast stream copy or re-encoded) | FFmpeg concat |
 | Side-by-Side Preview | Before/after frame comparison modal for effects | FFmpeg + base64 |
 | Watermark Removal | Remove logos via delogo or LaMA AI inpainting | FFmpeg / LaMA |
+| Color Match | Match the color profile of one clip to a reference clip using YCbCr histogram matching | OpenCV |
 
 ### AI & ML Tools
 
@@ -122,12 +128,14 @@ OpenCut v1.3.0 includes **142 API routes**, **6 panel tabs** with **43 sub-tabs*
 | Face Swap | Replace faces with reference image | InsightFace |
 | Style Transfer | Neural artistic style transfer | PyTorch models |
 | Auto Thumbnails | AI-scored frame extraction for thumbnails | OpenCV scoring |
+| Auto Zoom Keyframes | Face-detected push-in zoom for talking-head content. Returns keyframes for the Premiere Pro Motion effect | OpenCV + ExtendScript |
+| Multicam Auto-Switching | Speaker diarization → camera cut list. Maps speakers to track indices and generates multicam edits | pyannote.audio + ExtendScript |
 
 ### Export & Batch
 
 | Feature | Description | Engine |
 |---------|-------------|--------|
-| Platform Presets | YouTube, TikTok, Instagram, Twitter/X, LinkedIn, Podcast | FFmpeg encode |
+| Platform Presets | YouTube, TikTok, Instagram, Twitter/X, LinkedIn, Podcast, Snapchat Story, Facebook Reel, Facebook Post, YouTube Long Form, Pinterest Video, Podcast MP3 | FFmpeg encode |
 | Social Quick Export | One-click export optimized per platform's format and limits | FFmpeg |
 | Batch Processing | Process multiple clips in parallel | ThreadPool |
 | Transcript Export | SRT, VTT, ASS, plain text, timestamped | Built-in |
@@ -172,6 +180,60 @@ OpenCut v1.3.0 includes **142 API routes**, **6 panel tabs** with **43 sub-tabs*
 
 ---
 
+## Timeline Integration
+
+OpenCut v1.5.0 adds direct write-back to the active Premiere Pro sequence via ExtendScript:
+
+| Feature | Description |
+|---------|-------------|
+| Apply Cuts | Remove silences, repeated takes, or custom ranges directly from the timeline |
+| Beat Markers | Add detected beats as sequence markers at the click of a button |
+| Multicam Cuts | Apply speaker-driven cuts directly to the open sequence |
+| Clip Keyframes | Write scale/position keyframes to Motion effect for auto-zoom |
+| Batch Rename | Rename project panel clips with find/replace patterns |
+| Smart Bins | Auto-sort project items into bins by rule (name, type, duration) |
+| Export from Markers | Batch-export clip ranges defined by sequence markers |
+| Native Captions | Import SRT as a native Premiere Pro caption track |
+
+---
+
+## Search & AI Commands
+
+### Footage Search
+
+Index your entire media library by spoken content. Search across all clips by keyword or phrase. Results include clip path, timestamp, and matched text snippet.
+
+### Natural Language Commands
+
+Type commands in plain English — OpenCut maps them to API routes using keyword matching or LLM analysis:
+- *"remove silence from this clip"* → `/captions/silence`
+- *"add captions in Spanish"* → `/captions/transcribe` with language=Spanish
+- *"generate YouTube chapters"* → `/captions/chapters`
+
+### Post-Production Deliverables
+
+Generate professional post-production documents from sequence data:
+- **VFX Sheet** — Shot list with effects for the VFX supervisor
+- **ADR List** — Dialogue list for re-recording sessions
+- **Music Cue Sheet** — Music usage log for licensing/clearance
+- **Asset List** — Complete media inventory with usage counts
+
+---
+
+## UXP Panel
+
+A parallel panel (`com.opencut.uxp`) is available for Premiere Pro 25.6+ using Adobe's modern UXP platform:
+
+- **Modern JavaScript** — ES modules, async/await, native `fetch()`
+- **Same Python backend** — Connects to the same local server on port 5679
+- **Auto port detection** — Scans ports 5679–5689 automatically
+- **7 tabs** — Cut & Clean, Captions, Audio, Video, Timeline, Search, Deliverables
+- **Timeline API** — Uses the `premierepro` UXP module for direct sequence access (where available)
+
+> **Note:** The CEP panel (`com.opencut.panel`) remains the primary panel and supports Premiere Pro 2019+. The UXP panel is the future-facing option for 25.6+.
+
+---
+
 ## Architecture
 
 ```
@@ -235,6 +297,32 @@ pip install torch torchvision torchaudio --index-url https://download.pytorch.or
 | `OPENCUT_OUTPUT_DIR` | Source file dir | Default output directory |
 | `WHISPER_MODELS_DIR` | `~/.cache` | Whisper model cache |
 
+## CLI Usage
+
+```bash
+# Generate YouTube chapters
+opencut chapters interview.mp4 --provider ollama --model llama3
+
+# Detect repeated takes
+opencut repeat-detect recording.mp4 --threshold 0.6
+
+# Search footage library
+opencut search index *.mp4
+opencut search query "camera lens focal length"
+
+# Match colors to reference
+opencut color-match source.mp4 reference.mp4
+
+# Normalize loudness
+opencut loudness-match clip1.mp4 clip2.mp4 --target-lufs -14
+
+# Generate post-production documents
+opencut deliverables --sequence-json sequence.json --type all
+
+# Natural language editing
+opencut nlp "remove silence and add captions in Spanish" --file video.mp4
+```
+
 ## FAQ
 
 **Q: The panel says "Server offline"**

README.md

@@ -2,11 +2,11 @@
 <ExtensionManifest xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
                    Version="7.0"
                    ExtensionBundleId="com.opencut.panel"
-                   ExtensionBundleVersion="1.4.0"
+                   ExtensionBundleVersion="1.5.0"
                    ExtensionBundleName="OpenCut">
 
     <ExtensionList>
-        <Extension Id="com.opencut.panel.main" Version="1.4.0" />
+        <Extension Id="com.opencut.panel.main" Version="1.5.0" />
     </ExtensionList>
 
     <ExecutionEnvironment>