A macOS-native automation skill that generates Final Cut Pro rough-cut timeline variants from raw talking-head footage without cloud media uploads.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install roughcut-openclaw
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 roughcut-openclaw 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 RoughCut (macOS)?
RoughCut is a powerful local-first automation designed for video editors and content creators who want to streamline their post-production workflow. By utilizing Openclaw Skills, this tool runs headlessly on macOS to process raw talking-head video files, identifying and removing bad takes or irrelevant content. It bridges the gap between raw footage and the editing suite by producing a set of ready-to-import FCPXML files.
The primary value of this skill lies in its privacy and efficiency. Since all heavy lifting occurs locally on your machine using FFmpeg and Python, no media is ever uploaded to a third-party server. It allows editors to bypass the time-consuming process of the first cut, enabling them to focus immediately on the creative aspects of their project within Final Cut Pro.
RoughCut (macOS) Use Cases
- Automated identification and removal of bad takes in talking-head footage.
- AI-driven fluff and off-topic content removal based on specific video goals.
- Generating multiple timeline variants to experiment with different edit densities.
- Processing remote footage from S3 or pre-signed URLs via local download and headless execution.
How RoughCut (macOS) Works
- The agent locates the local video file or downloads it from a provided HTTPS URL into a temporary input directory.
- User preferences are checked to determine if bad takes or off-topic fluff should be removed from the final sequence.
- The RoughCut shell script is invoked using the configured repository root and output directories.
- If fluff removal is enabled, the Gemini API is leveraged to analyze the transcript and filter content based on the provided topic.
- The tool processes the video locally using FFmpeg to calculate timings and generate metadata.
- A compressed ZIP file containing several .fcpxml variants is produced, which the user can import directly into Final Cut Pro.
RoughCut (macOS) Setup
To use this skill, ensure your macOS environment has bash, python3, curl, ffmpeg, node, and npm installed. You must also clone the RoughCut repository and configure your Openclaw Skills environment with the correct repo_root and output_root paths. If you intend to use the fluff removal feature, a GEMINI_API_KEY must be exported in your environment variables.
bash "$REPO_ROOT/scripts/openclaw/roughcut.sh" \
--video "/Users/username/Movies/raw_footage.mp4" \
--out "$OUTPUT_ROOT" \
--remove-bad-takes true \
--remove-fluff false
RoughCut (macOS) Data Schema & Taxonomy
The skill organizes its output within the defined output_root to ensure compatibility with other Openclaw Skills. Data is structured as follows:
| Component | Location | Description |
|---|---|---|
| Input Media | RoughCut.inputs/ |
Stores videos downloaded via URL or temporary working copies. |
| XML Variants | RoughCut.xml_variants.zip |
The primary output containing multiple .fcpxml files for import. |
| Debug Data | debug_zip |
Generated only on failure; contains logs and intermediate processing state. |
| Result Metadata | Standard Output | A single-line JSON object detailing the paths to the generated assets. |
RoughCut (macOS) Advanced Features
- Support for direct download URLs with SHA256 integrity verification for secure remote media processing.
- Multi-variant FCPXML generation providing different levels of edit aggressiveness.
- Integration with the Gemini API for intelligent, topic-aware content pruning.
- Headless macOS execution allowing for background batch processing of long-form footage.
- Local-first architecture ensuring high-resolution media never leaves the user's workstation.
SKILL.md
Loading
Related Openclaw Skills
Reader Deep Dive
A sophisticated intelligence briefing skill that bridges the gap between your current reading list and your historical knowledge archive.
Oura Ring Integration
A Python-powered skill for fetching Oura Ring sleep, readiness, and resilience data through the Oura V2 API.
ChatGPT History Import
A utility to migrate ChatGPT conversation history into OpenClaw's memory search by converting exports into searchable SQLite vector databases.
MemePickup Wingman
An AI-powered dating assistant that generates personalized openers, conversation replies, and profile analysis using the MemePickup API.
Ironclaw
Ironclaw is a real-time safety and threat classification engine designed to protect AI agents from malicious content and unauthorized data exfiltration.
Groq Completion
A high-performance integration that leverages Groq's LPU technology for near-instantaneous AI text generation and completions.
Featured*
