How does clipit differ from simply asking an AI agent to edit a video for me?

clipit provides the precise frame-by-frame timing and FFmpeg command generation needed for professional effects like speed ramping and beat-syncing, which standard LLMs cannot calculate or execute on their own.

What types of video sources and languages does the skill support?

The skill supports YouTube links and direct video file uploads, and it includes specialized models for both English and Hindi transcription.

What is included in the purchase of this skill?

You will receive the complete agentic workflow, a pre-configured Python environment spec including FFmpeg and Whisper integrations, and the proprietary JSON story orchestration logic.

What are the technical requirements for the initial installation?

The initial setup requires a one-time 700MB download for the core engine, with an additional 3GB optional download if you require high-accuracy Hindi transcription.

How does the agent ensure the subject stays in frame during the 9:16 crop?

clipit uses automated face-safe framing and focus-hunting cuts to ensure that even when converting wide 16:9 footage to vertical 9:16, the main subject remains perfectly centered and visible.

ClipIt

Turn long-form videos and YouTube links into beat-synced 9:16 shorts, reels, and high-end velocity edits.

Free

Security scanned

Works with the AI Agents you already use

+22 more

+18 agents

ClipIt

Example session with this skill installed

I have this link to the cricket match highlights: https://youtu.be/example. Use the song "Night Drive" I uploaded. Make a punchy 15-second velocity edit of the winning six. Level 8.

Read your context and instructions
Compiled the clipit

I've analyzed the highlights and the track. I picked the final over sequence, specifically the winning shot at 09:42, and synced the impact to the drop at 0:04 of your song. I applied a level 8 punchy edit with zoom-cuts, a teal-orange look, and motion-blur jolts on the beat.

Video Edit Summary

Metric	Value
Duration	14.8 seconds
Edit Level	8 (Punchy)
Format	9:16 (Vertical)
Framing	Face-safe center crop with mirror-edge fill
Music Sync	128 BPM, drop at 04.2s


# Render command used for the final output
PY="$HOME/.clipit/venv/bin/python"; CLIP="$SKILL_DIR/clip.py"
"$PY" "$CLIP" edit ./input/match_highlights.mp4 --music ./music/night_drive.mp3 --length 15 --level 8 --hook-start 582 --hook-end 597 --look teal_orange --flow

The video has been rendered to ./output/cricket_velocity_8.mp4. I checked the 30-frame contact sheet; the framing correctly follows the batsman throughout the swing and the motion-blur scales naturally with the speed ramps.

Next steps

Review the generated cricket_velocity_8_sheet.jpg to confirm the color grade and framing.
Run with --level 5 if the shakes on the beat are too aggressive for your preference.
Use postxport.com to share the final MP4 for review or upload.

Connects securely to your tools. The creator never sees your data.

About this skill

High-quality video editing requires thousands of manual keyframes and perfect timing. This skill automates the labor of turning long-form footage into viral 9:16 shorts, reels, and fan edits while maintaining professional "mogged" aesthetics.

What it does

Viral clipping finds the strongest hook in a transcript and cuts dialogue-driven shorts with synchronized captions.
Velocity editing applies optical-flow slow motion and speed ramps synced precisely to a song's beats and drops.
Story orchestration builds complex narrative arcs using a JSON spec to manage text cards, dialogue intros, and montage payoffs.
Visual engineering handles face-safe 9:16 framing, focus-hunting cuts, motion-blur, and automated color grading via LUT matching.
Format versatility generates comeback reels, split-screen podcast clips, and hyper-edits with levels from clean to high-intensity.

How it works

Intake where the agent gathers YouTube links or files and analyzes the song for tempo and drop candidates.
Analysis where it sheets the footage to find hero shots and transcribes speech for dialogue snapping.
Execution where it writes a story spec or runs CLI commands like edit or talk based on the desired edit level.
Refinement where it reviews contact sheets, checks for still-frame violations, and iterates on pacing.

Frameworks & tools

The system uses a custom Python environment including ffmpeg, yt-dlp, and OpenAI Whisper. It supports specialized extensions for Hindi speech and AI subject cut-outs.

Why this beats prompting it yourself

Standard prompts cannot handle frame-accurate beat matching or complex video processing pipelines. This skill bridges the gap between LLM reasoning and professional video rendering, ensuring your agent actually outputs a production-ready MP4 rather than just a description of one.

Use cases

Convert long-form podcast episodes into captioned TikTok clips.
Generate high-energy sports or gaming highlights synced to phonk music.
Create "Comeback" or "Tribute" story reels with emotional dialogue intros.
Recreate specific After Effects or CapCut visual styles from reference videos.

Known limitations

Initial setup requires a one-time 700MB download. Hindi transcription requires an optional 3GB model. Links and large model downloads are restricted when running in the Claude.ai sandbox.

How to use

Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.

Reviews

No reviews yet

Be one of the first to try it. Every listed skill passes our trust checks below.

Security scanned

Passed our 8-point scan before listing

Fresh listing

Recently published to Agensi

Free forever

No account required to browse

Trust & safety

Security scanned

Verified clean 7 days ago

Free to download with an account

Requires

Googleapis

Python

macOS

Listed7 days ago

ClipIt