skills.rest
openakita
openakitaCOMMUNITY
Explore all skills by this author →
GitHub

footage-gate

Review source media, cut silence, auto color grade, and run cut-boundary QC via local FFmpeg pipelines.

2.0k|274|Updated Jan 30, 2026
Search Tags:#FFmpeg#post-production#color grading#edl#silence removal#hdr tonemapping#video qc
One-click install
npx skills add https://github.com/openakita/openakita --skill footage-gate
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: footage-gate
Source: https://github.com/openakita/openakita/tree/main/plugins/footage-gate
Command: npx skills add https://github.com/openakita/openakita --skill footage-gate

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires numpy, Pillow.

What problem does it solve?

Raw footage and exported masters often ship with hidden defects — low resolution, dead air, flat color, HDR clipping, or bad cut boundaries — and manually checking each one is slow and error-prone. Footage Gate automates post-production quality control with four deterministic, fully local FFmpeg pipelines, so you can verify and fix media before delivery without any cloud dependency.

Core Features & Use Cases

  • Source Review: Probe video, audio, and image files to flag low resolution, mono audio, or too-short clips, with an optional DashScope Paraformer transcription summary.
  • Silence Cut: Detect and remove dead air using pure-NumPy RMS analysis (no aubio dependency), then concat the kept segments with FFmpeg.
  • Auto Color Grade: Sample frames with signalstats to derive a clamped eq filter chain, automatically prepending an HDR-to-SDR tonemap chain for HLG/PQ sources.
  • Cut Boundary QC: Run four checks (boundary frame jitter, waveform spikes, subtitle safe zones, EDL duration consistency) against an EDL JSON, with an optional auto-remux repair loop of up to 3 attempts.
  • Use Case: After exporting a master cut, run cut_qc with your EDL to catch a subtitle hidden behind the TikTok bottom UI and an audio spike at a cut point, then let the auto-remux loop fix both before delivery.

Quick Start

Ask the agent to create a footage-gate task, for example: run a source review on my uploaded video file and show me the quality risk report.

Frequently Asked Questions about footage-gate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I remove silence from a video automatically?▼

Use the silence_cut mode, which detects non-silent intervals with pure-NumPy RMS analysis and concatenates the kept segments via FFmpeg. You can tune threshold_db, min_silence_len, min_sound_len, and pad parameters per task.

How to check video cut quality against an EDL file?▼

Use the cut_qc mode with an EDL JSON payload describing your cuts. It runs four checks — boundary frame jitter, waveform spikes, subtitle safe zones, and duration consistency — and optionally auto-remuxes up to 3 times to fix issues.

Does this video QC tool work without an internet connection?▼

Yes, all four modes run entirely on local FFmpeg with no LLM or API dependency in the default path. The only optional cloud call is DashScope Paraformer transcription in source_review, which is off by default and requires your own API key.

Can FFmpeg auto color grade HDR footage?▼

Yes, the auto_color mode detects HDR transfers (smpte2084, arib-std-b67) via ffprobe and prepends a zscale plus tonemap=hable chain before the eq grade filter. This converts HLG/PQ sources to BT.709 SDR and prevents the clipping that occurs when grading HDR directly.

What are the limitations of automated cut boundary QC?▼

The boundary frame check uses histogram-style pixel diffs, so hard-edged whip-pans can register as false positives. Auto color only applies a single global eq chain, and scene-aware grading is deferred to a later version.

Why does silence detection fail with aubio on newer Python?▼

The aubio package fails to install on Python 3.10+ with NumPy 1.24+, which is a known upstream issue. This skill avoids the problem entirely by implementing silence detection in pure NumPy with no aubio, madmom, or librosa dependency.

Related Skills

Explore More Design & Creative Skills→
broomvabroomva
3

video-cut

Edit raw video footage into finished cuts using local transcription and EDL-based rendering.

Community
Advanced
citedycitedy
31

citedy-video-shorts

Generate and publish branded AI avatar lip-sync video shorts to Instagram Reels and YouTube Shorts.

Official
Advanced
NoizAINoizAI
526

video-translation

Translate and dub spoken audio in videos using subtitle files and timeline-based TTS.

Official
Intermediate
damionrashforddamionrashford
15

workflow-editorial-interchange

Translate timelines between NLEs using OTIO as the pivot.

Community
Advanced