Automates video editing for voice-over content. Uses semantic understanding to detect and remove errors like repetitions and corrections. Integrates with custom dictionaries for accurate transcription of technical terms.
git clone https://github.com/Ceeon/videocut-skills.githttps://github.com/Ceeon/videocut-skills
[{"step":"Prepare your inputs: Upload the raw video file and provide a technical dictionary (if applicable) in a text file or as a list of terms. Specify your target duration and preferred output format (e.g., MP4, MOV) in the prompt.","tip":"Use tools like Otter.ai or Descript to generate an initial transcript if you don’t have one. For technical terms, include variations (e.g., 'AI' vs. 'A.I.') to improve accuracy."},{"step":"Customize the prompt: Replace [VIDEO_FILE], [TECHNICAL_TERMS], [TARGET_DURATION], and [FORMAT] with your specific requirements. Include additional constraints like 'preserve humor' or 'keep pauses under 0.8s' if needed.","tip":"For YouTube content, aim for a target duration 10-15% shorter than the original to improve engagement. Use the 'smooth transitions' parameter sparingly to avoid over-editing."},{"step":"Run the AI tool: Paste the prompt into your AI assistant (e.g., Claude, ChatGPT) or a specialized video editing tool like Runway ML or Pictory. Attach the video file and dictionary if required.","tip":"For large files, compress the video first (e.g., using HandBrake) to reduce processing time. If the AI tool supports it, enable 'semantic editing mode' for better context-aware cuts."},{"step":"Review and refine: Check the AI’s edit log and output video. Use editing software (e.g., Adobe Premiere, CapCut) to manually adjust cuts, transitions, or subtitles if needed.","tip":"Focus on preserving the speaker’s tone and intent. If the AI removes too much natural speech, tweak the prompt to retain more pauses or emphasis."},{"step":"Export and distribute: Finalize the video with your preferred settings (e.g., aspect ratio, captions) and publish it to your target platform.","tip":"For social media, create a 15-30 second highlight reel from the edited video using tools like Canva or Adobe Express."}]
Automatically detect and correct speech errors in video recordings.
Generate high-quality subtitles for videos using a custom dictionary.
Streamline the video editing process by identifying silent segments and repetitions.
Facilitate quick reviews of video content through an AI-assisted auditing interface.
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/Ceeon/videocut-skillsCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Edit the video [VIDEO_FILE] for a voice-over script using the following requirements: 1. Remove all filler words (e.g., 'um', 'uh', 'like') and false starts. 2. Correct grammatical errors in the spoken content while preserving natural speech flow. 3. Use the provided technical dictionary [TECHNICAL_TERMS] for accurate transcription of industry-specific terms. 4. Ensure the final cut is under [TARGET_DURATION] minutes with smooth transitions. 5. Export the video in [FORMAT] with subtitles burned in. Include a timestamped log of all edits made.
Here’s the edited version of your video, *Introduction to Quantum Computing.mp4* (original duration: 12:47, final cut: 8:23). The AI identified and removed 17 filler words, corrected 3 grammatical errors (e.g., 'we was' → 'we were'), and standardized 8 technical terms using your provided dictionary (e.g., 'qubit' was corrected from 'cubit' in two instances). The video now features seamless cuts between segments, with transitions added at the 2:15, 4:30, and 6:45 marks to improve pacing. The final export is in 1080p MP4 with burned-in subtitles, and the timestamped edit log is attached below: **Edit Log:** - 00:01:22 - Removed 'um' and corrected 'we was' to 'we were'. - 00:03:45 - Standardized 'qubit' (previously misheard as 'cubit'). - 00:05:10 - Removed false start 'So, like, let's talk about...' → 'Let’s discuss...'. - 00:07:33 - Adjusted audio level to -12dB for consistency. - 00:08:10 - Added crossfade transition (0.5s) between segments. The subtitles now reflect the corrected transcript, with technical terms highlighted in bold for emphasis. The video maintains its original educational tone while being more concise and professional. Would you like to review the subtitles or adjust any transitions further?
AI assistant built for thoughtful, nuanced conversation
IronCalc is a spreadsheet engine and ecosystem
ITIL-aligned IT service management platform
Customer feedback management made simple
Enterprise workflow automation and service management platform
Automate your spreadsheet tasks with AI power
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan