An automated utility designed to detect high-energy moments in stream recordings and slice them into highlight videos.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install skill-live-highlight-slicer
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install skill-live-highlight-slicer using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Live Highlight Slicer?
The Live Highlight Slicer skill enables automatic segmentation of live stream video recordings such as .webm, .mp4, or .mov files. This tool identifies critical video highlights using audio intensity thresholds, visual scene transitions, or keyword density from ASR transcripts. This replaces the repetitive process of manually reviewing recordings and clipping files by hand.
By implementing this skill within Openclaw Skills integrations, developers can automate post-stream content generation, creating quality summary clips or combined highlights with minimized processing overhead.
Live Highlight Slicer Use Cases
- Isolating highlight clips from gaming or broadcasting sessions using audio amplitude surges.
- Detecting scene changes to capture transitions such as product switches or presenter changes.
- Evaluating ASR transcript files to extract segments containing specific keywords.
- Creating social media promotional clips from longer recordings under the Openclaw Skills framework.
- Compiling all isolated highlight clips into a single consolidated output video.
How Live Highlight Slicer Works
- Parameter Validation: Ensures the presence of required CLI parameters, validates the local video path, and checks dependencies.
- Feature Extraction: According to the chosen method (
audio,scene,hybrid,combined, orasr), it analyzes standard deviation peaks, ffmpeg scene changes, or transcript windows. - Consolidation of Clips: Merges timeline intervals that are closer than 3 seconds apart to avoid choppy video transitions.
- Duration Filtering: Removes any candidate segments that fall below the minimum clip duration limit.
- Segment Generation: Pads remaining segments and triggers ffmpeg tasks to output individual video clips.
- Compilation Video: If enabled, creates concat file lists to produce a single merged highlight presentation.
Live Highlight Slicer Setup
Ensure you have Python >= 3.9, along with ffmpeg and ffprobe binaries in your system path.
First, install the necessary Python dependencies:
pip install librosa numpy pydub
Run the script directly using Python, providing the absolute path to your video file:
python3 skill-live-highlight-slicer/scripts/highlight_slicer.py --input "/absolute/path/live-recording.mp4"
Modify behavior by specifying alternate parameters like --method scene or passing --asr-file for text-based analysis.
Live Highlight Slicer Data Schema & Taxonomy
CLI Argument Schema
| Parameter | Required | Default | Description |
|---|---|---|---|
--input |
Yes | None | Absolute path to the source video recording |
--method |
No | hybrid |
Analysis method: audio, scene, hybrid, combined, asr |
--threshold |
No | 1.5 |
Standard deviation multiplier for audio peak threshold |
--scene-threshold |
No | 0.1 |
Scene transition threshold for visual change detection |
--min-clip-duration |
No | 5 |
Minimum length of generated clips in seconds |
--padding |
No | 2 |
Duration in seconds padded to the start and end of clips |
--output-dir |
No | ./highlights |
Target output directory for files |
--merge |
No | true |
Boolean flag to generate a single combined video of highlights |
--asr-file |
No | None | File path to local ASR transcript |
--asr-window |
No | 8 |
Size of sliding window in seconds for keyword analysis |
--top-n |
No | 8 |
Target maximum number of segments to export in ASR mode |
Generated Output Artifacts
All generated content is saved inside the specified directory:
clip_01_[start]_[end].mp4: Segmented video files of detected highlight events.segments.json: Structured log containing precise start/end timestamps, extraction methods, relative metrics, and index details.merged_highlights.mp4: Merged compiled highlight reel (generated only when--merge=true).
Live Highlight Slicer Advanced Features
- Flexible Hybrid Strategy: Leverages temporal intersections of audio energy spikes and visual scene cuts to yield accurate summary selections.
- ASR Keyword Integration: Scans structured JSON transcripts (using keys like text, start, end) to pinpoint segments by specific terms or call-to-actions.
- Custom Padding & Merging: Applies padding buffers to segments to prevent clipped audio and merges close intervals to preserve contextual continuity.
- Predictable Exit Protocol: Employs precise exit codes (0 to 3) and log levels to interface predictably with other tools in Openclaw Skills automation pipelines.
SKILL.md
Loading
Related Openclaw Skills
File Hash Verifier
Verify files with SHA-256 checksums to detect corruption, tampering, and release drift in seconds.
draw.io Diagrammer Skill
An advanced AI-agent skill for programmatic generation, styling, and visual validation of draw.io diagrams.
Software Engineering Discipline
A rigorous technical framework that calibrates architectural trade-offs and enforces strict code verification boundaries for AI coding agents.
Draw.io Diagrammer Skill
An automated agent skill designed to generate, export, and visually inspect draw.io diagrams for technical documentation.
Douyin Creator Center Video Upload
An automation tool that uploads local video files to the Douyin Creator Center and manages the publication workflow via a Chrome DevTools Protocol connection.
OppHub OPC Skill
An automation and integration client for OppHub OPC matchmaking, designed with zero interactive prompts for autonomous AI agents.
Featured*
