- Home
- Skills
- Workflow Automation
- Dead Air Cutter
Works with the AI Agents you already use
Dead Air Cutter
Remove silences from videos and podcasts with automated ffmpeg cutting and synced subtitle generation.
$10
Dead Air Cutter
Example session with this skill installed
Cut the dead air out of interview_raw.mp4 using the gentle style. I have the MCP server ready.
- Read your context and instructions
- Compiled the dead air cutter
I have completed the dead-air removal process for interview_raw.mp4. The audio was split into 3 parts and transcribed. I applied the
gentle style, which preserves 0.3s of air around speech for a natural flow.
Processing Summary
| Phase | Status | Details |
|---|---|---|
| Analysis | Success | Found 42 silence segments |
| Transcription | Success | 3 parts processed ($0.152 total) |
| Rendering | Success | H.264 CRF 18 / AAC 192k |
| Original Length | 12:45 | - |
| New Length | 10:22 | 2:23 removed |
Files Generated
The following files have been created in the source directory:
interview_raw_cut.mp4: The edited video file with silences removed.interview_raw_cut.srt: Synchronized subtitles for the edited version.
A temporary work directory .dac-interview_raw/ was used for transcripts and manifests. The transcripts are saved locally, so you can re-run the render with a different style (e.g., jump) without paying for transcription again.
Next steps
- Review
interview_raw_cut.mp4to ensure the 0.3s padding is sufficient for your needs. - If the cuts are too slow, run the
rendercommand again with thenormalstyle. - Open the
.srtfile in a player like VLC to verify subtitle sync.
Connects securely to your tools. The creator never sees your data.
About this skill
Editing a raw talking-head video or podcast often involves hours of tedious manual cutting to remove silences and pauses. This skill automates the process of tightening recordings, creating jump cuts, and generating synchronized subtitles. It handles the heavy lifting of audio transcription and precise timestamping so you can skip the manual timeline scrubbing.
What it does
- Silence detection identifies dead air throughout the recording using automated transcription timings.
- Automated cutting executes precise edits using
ffmpegto remove pauses based on your chosen style. - Subtitle synchronization generates
.srtfiles that perfectly match the new, shorter duration of the cut video. - Adjustable pacing offers gentle, normal, or jump-cut modes to control the aggressiveness of the edits.
- Cost estimation calculates transcription fees before any spending occurs to keep costs predictable.
How it works
- Split: The tool analyzes the local video file, splits the audio into parts, and provides a total transcription cost estimate.
- Transcribe: The agent calls the
transcribe-mcptool for each audio part to generate precise speech timestamps. - Merge and Analyze: Timings are processed locally to map out every silence that needs to be removed.
- Render: The agent runs the final
ffmpegcommand to produce the cut video and its matching subtitle file.
Frameworks & tools
This skill requires ffmpeg and ffprobe installed on the system path. It utilizes a local Node.js script for orchestration and requires the transcribe-mcp server (v0.3.0+) for cloud-based transcription.
Why this beats prompting it yourself
Writing manual ffmpeg filter strings for dozens of individual cuts is prone to syntax errors and audio desync. This skill manages the complex math of shifting subtitle timestamps and re-encoding video streams without manual intervention.
Use cases
- Tighten up long-form podcast recordings by removing awkward pauses.
- Create aggressive jump-cuts for YouTube vlogs or social media clips.
- Generate synced subtitles for edited lectures or tutorials.
- Preview different editing styles (gentle vs. aggressive) without re-transcribing.
Known limitations
Requires a dedicated wallet with USDC on Base for transcription payments. Supports video re-encoding to H.264 only. Individual audio parts are capped at 800 KB for processing.
Requires ffmpeg, Node 20+, and a funded USDC wallet on Base. Re-encodes to H.264/AAC. Processing cap of 800 KB per audio part.
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
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 27 days ago
Requires