- Home
- Skills
- Social Media
- Video Build by shash.builds
Video Build by shash.builds
A script file, voice and face clip to a finished 1080x1920 reel: shots on a timeline, word-by-word brand captions, music bed, -14 LUFS. Free and local.
Free
Works with the AI Agents you already use
Video Build by shash.builds
Example session with this skill installed
python video_build.py video.md -o day-01.mp4
- Read your context and instructions
- Compiled the video build by
day-01.mp4 1080x1920 30 fps -14 LUFS
Connects securely to your tools. The creator never sees your data.
About this skill
Make the daily video in one command instead of an hour in an editor.
You write a short video.md: where the voice and face clip are, and a shot list (at 0:04 show this screenshot with this line of text). The skill transcribes the voice locally, cuts every shot to the timeline, burns in word-by-word captions with the current word highlighted, adds each shot's on-screen text, lays an optional music bed under the voice and sets loudness to -14 LUFS, the level Instagram, TikTok and YouTube expect.
What you get:
- A 1080x1920, 30 fps MP4 ready to post as a reel, short or TikTok
- Shots: your face clip, a screenshot or screen recording (stills get a slow push-in), or a title card
- Captions in a clean brand style: cream words on a dark box, the spoken word in orange
- An optional closing stinger clip and a credits file for the music
Before rendering, a dry run prints every shot with its start and end, so you catch mistakes early.
Requirements: ffmpeg, Python with Pillow, and the bundled transcribe tool for captions (or bring your own word timings). Everything runs on your computer, nothing is uploaded.
Free and MIT licensed. Part of the shash.builds Toolbox.
The face clip plays from the same moment as the voice, so they must be recorded together or lined up first. Captions can contain misheard words; fix them by editing the words.json and rendering again. Only use music you have the rights to. Tested on Linux; not tested on Windows yet.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 13 days ago
- Free with an Agensi account
Requires
Permissions
Allowed Hosts
File Scopes
Runs locally; nothing is uploaded. Brand fonts download once from cdn.jsdelivr.net. For captions, faster-whisper may be installed from PyPI and a Whisper model downloaded on first run.