메뉴
Night Grind Radio - Lo-fi video production pipeline using direct ffmpeg (bypasses Remotion disk issues)
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Run a repeatable codebase polish pass. Use when the user says 'polish', 'cleanup', 'tidy', 'housekeeping', or 'shine'. Runs all audit scripts, identifies safe fixes, implements them, and reports a before/after scorecard. Safe to run repeatedly — each pass finds diminishing returns.
Monitor Nostr for RUNSTR mentions, bug reports, and community sentiment. Use when the user says 'vibes', 'check mentions', 'what are people saying', 'any bug reports', 'community pulse', or 'check nostr'. Can be looped with /loop for recurring monitoring.
Create a RUNSTR pre-release. Use when the user says "prerelease", "pre-release", "create a release", "build release", "release the app", or "ship it". Handles changelog, version bumps, APK build, and GitHub release creation.
Give your AI agent access to your health and fitness data from RUNSTR. Fetches workouts, habits, journal entries, mood, steps, competition leaderboards, and more from Nostr. Use when the user asks about their workouts, fitness history, health habits, mood tracking, competition rankings, who's winning, what place they're in, or wants AI fitness coaching based on real data.
Manage the RUNSTR iOS simulator workflow. Use when the user says "run the app", "start the simulator", "reload", "restart metro", or needs help with the iOS development environment. Handles Metro bundler, Xcode builds, and simulator control.
| name | lofi-video-render |
| description | Night Grind Radio - Lo-fi video production pipeline using direct ffmpeg (bypasses Remotion disk issues) |
| metadata | {"tags":"ffmpeg, video, lofi, youtube, night-grind-radio, render"} |
Use this skill whenever producing a Night Grind Radio lo-fi video. This includes:
Remotion renders via headless Chromium, creating a webpack bundle (~200MB), Chromium profile, and temp frames that consume 3-5GB+ of disk for a 79-minute video. On a disk-constrained machine (228GB, ~95% full), this repeatedly fails with ENOSPC.
Direct ffmpeg streams the encode with near-zero temp disk usage. Only the growing output MP4 needs space (~1-1.5GB for 79 min at 2Mbps).
Read the rule files for each step of the pipeline:
