mixcut
Create short mixed-cut videos from YouTube clips with automated analysis, clipping, and professional subtitle processing.
Best use case
mixcut is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Create short mixed-cut videos from YouTube clips with automated analysis, clipping, and professional subtitle processing.
Teams using mixcut should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/MixCut-Skill/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How mixcut Compares
| Feature / Agent | mixcut | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Create short mixed-cut videos from YouTube clips with automated analysis, clipping, and professional subtitle processing.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# MixCut Skill
## CRITICAL INSTRUCTION: MANDATORY CONFIGURATION
You MUST NOT proceed to downloading or processing until the Project Configuration is fully defined.
**Step 1: The Configuration Gate**
Check if the user has provided ALL of the following:
- **Video Title**: Main title for intro and segment cards.
- **Caption Language**: Target language for subtitles (e.g., English, etc.). *Crucial: No bilingual output.*
- **Aspect Ratio**: 16:9 (Desktop), 9:16 (Shorts/TikTok), or 1:1.
- **Target Duration**: 30s, 60s, 120s, or 180s.
- **Pace**: Fast, Medium, or Slow.
- **Style**: Highlight Montage or Beat-cut.
- **Transitions**: none, fade, slide, flash, or glitch.
If ANY of these are missing, you MUST use the `AskUserQuestion` tool to obtain them.
DO NOT assume defaults.
DO NOT start downloading until configuration is complete.
---
## Workflow
### Phase 1: Sources & Workspace Setup
1. **Gather Sources**: Ask for YouTube URLs from the user.
2. **Setup Workspace**:
- Ensure the directory `./work` exists in the project root.
- Prepare `./work/assets` and `./work/outputs` subdirectories.
### Phase 2: Content Analysis
For each YouTube URL provided:
1. **Download**:
- Run `python skills/Youtube-clipper-skill/scripts/download_video.py <url> "./work/assets/youtube"`
2. **Analyze**:
- Run `python skills/Youtube-clipper-skill/scripts/analyze_subtitles.py <subtitle_path>`
3. **Selection**:
- Present analyzed chapters/highlights to the user.
- Confirm which segments to include in the final cut.
### Phase 3: Assembly & Processing
1. **Configuration**: Create `./work/mixcut_config.json` with global settings and source clip metadata (start, end, title).
2. **Processing**:
- Run `python skills/mixcut/scripts/process_mixcut.py "./work/mixcut_config.json"`
- This script automates clipping, media normalization (ffmpeg), and generates the initial `timeline.json` and `final.srt`.
### Phase 4: Professional Subtitle Cleaning (Agent-Driven)
*This phase ensures high-quality, professional subtitles.*
1. **Read Timeline**: Access `./work/timeline.json`.
2. **Subtitle Optimization**:
Iterate through all subtitle entries and apply the following cleaning logic:
- **Remove Noise**: Strip speaker markers (e.g., `>>`, `>>>`) and sound cues (e.g., `[Music]`, `[Applause]`).
- **Deduplication**: Remove repeated words, stutters, and verbal fillers ("uh", "um", "you know").
- **Flow Improvement**: Merge fragmented phrases into natural, readable sentences.
- **Professional Polish**: Refine text into the target language using appropriate technical terminology.
3. **Update**: Save the optimized content back to `./work/timeline.json`.
4. **Synchronize**: Run `python skills/mixcut/scripts/process_mixcut.py "./work/mixcut_config.json" --action generate_srt` to sync changes to `final.srt`.
### Phase 5: Rendering
1. **Delegate to Remotion**:
- Use the `remotion` skill to render the project located in `skills/mixcut/remotion`.
- Provide the absolute path to `./work/timeline.json` as input props.
- Set output to `./work/outputs/final_mixcut.mp4`.
### Phase 6: Delivery Guard & Verification
*Mandatory verification before completion.*
1. **Integrity Check**:
- Verify `./work/outputs/final_mixcut.mp4` exists and has a file size > 0.
- Verify `./work/outputs/final.srt` exists and has a file size > 0.
2. **Troubleshooting**: If files are missing or empty, identify the failure point in previous phases and re-run the necessary steps.
### Phase 7: Completion & Cleanup
1. **Report**: Provide the paths to the final video and subtitle files.
2. **Retention Policy**: Do NOT delete the `./work` directory until the user explicitly confirms the results or you have verified the outputs physically.
---
## Technical Guidelines
- **Path Management**: Always use absolute paths when delegating tasks to external skills.
- **Cross-Platform**: Avoid hardcoding drive letters; use system-agnostic path resolution.
- **Performance**: Use parallel tool calls for independent download/analysis tasks where supported.Related Skills
grail-miner
This skill assists in setting up, managing, and optimizing Grail miners on Bittensor Subnet 81, handling tasks like environment configuration, R2 storage, model checkpoint management, and performance tuning.
ux
This AI agent skill provides comprehensive guidance for creating professional and insightful User Experience (UX) designs, covering user research, information architecture, interaction design, visual guidance, and usability evaluation. It aims to produce actionable, user-centered solutions that avoid generic AI aesthetics.
astro
This skill provides essential Astro framework patterns, focusing on server-side rendering (SSR), static site generation (SSG), middleware, and TypeScript best practices. It helps AI agents implement secure authentication, manage API routes, and debug rendering behaviors within Astro projects.
modal-deployment
Run Python code in the cloud with serverless containers, GPUs, and autoscaling using Modal. This skill enables agents to generate code for deploying ML models, running batch jobs, serving APIs, and scaling compute-intensive workloads.
vly-money
Generate crypto payment links for supported tokens and networks, manage access to X402 payment-protected content, and provide direct access to the vly.money wallet interface.
tech-blog
Generates comprehensive technical blog posts, offering detailed explanations of system internals, architecture, and implementation, either through source code analysis or document-driven research.
ontopo
An AI agent skill to search for Israeli restaurants, check table availability, view menus, and retrieve booking links via the Ontopo platform, acting as an unofficial interface to its data.
whisper-transcribe
Transcribes audio and video files to text using OpenAI's Whisper CLI, enhanced with contextual grounding from local markdown files for improved accuracy.
thor-skills
An entry point and router for AI agents to manage various THOR-related cybersecurity tasks, including running scans, analyzing logs, troubleshooting, and maintenance.
chrome-debug
This skill empowers AI agents to debug web applications and inspect browser behavior using the Chrome DevTools Protocol (CDP), offering both collaborative (headful) and automated (headless) modes.
lets-go-rss
A lightweight, full-platform RSS subscription manager that aggregates content from YouTube, Vimeo, Behance, Twitter/X, and Chinese platforms like Bilibili, Weibo, and Douyin, featuring deduplication and AI smart classification.
advanced-skill-creator
Meta-skill that generates domain-specific skills using advanced reasoning techniques. PROACTIVELY activate for: (1) Create/build/make skills, (2) Generate expert panels for any domain, (3) Design evaluation frameworks, (4) Create research workflows, (5) Structure complex multi-step processes, (6) Instantiate templates with parameters. Triggers: "create a skill for", "build evaluation for", "design workflow for", "generate expert panel for", "how should I approach [complex task]", "create skill", "new skill for", "skill template", "generate skill"