JPeetz/agent-skills / skills/marketing/local-video-production-m5
local-video-production-m5
VerifiedUse this skill when producing local videos on the 16GB M5 executor.
| File | Kind | Size |
|---|---|---|
| CHANGELOG.md | markdown | 586 B |
| SKILL.md | skill md | 9.6 KB |
| evals/evals.json | config | 2.4 KB |
| references/comfyui-mps-custom-node-pitfalls.md | markdown | 3.4 KB |
| references/kokoro-tts-mature-voice.md | markdown | 2.1 KB |
| references/ltx-mlx-a2v-setup.md | markdown | 3.9 KB |
| scripts/m5_env_check.shexecutable | script | 665 B |
evals/evals.json
View on GitHub{
"skill": "local-video-production-m5",
"version": "1.0.0",
"cases": [
{
"id": "e1",
"description": "Should trigger: local talking-video production on M5",
"input": "Produce a 'character presents a message' video locally on the M5.",
"shouldTrigger": true,
"expect": "Runs the local pipeline: Kokoro TTS, LTX-2.3 MLX A2V render, karaoke, trim, QA",
"assertions": [
"Studies existing on-disk reference pipeline first (herald_pipeline.sh + build_final_video.py)",
"Uses Q4 quant 512x512 low_ram config (not Q8 which swap-thrashes on 16GB)",
"Sets PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0",
"Shuts down LM Studio before renders",
"Quotes honest speed numbers (16 s/frame) when asked for ETA"
]
},
{
"id": "e2",
"description": "Should trigger: memory-constrained A2V rendering",
"input": "Render a 5-second A2V clip with LTX on this 16GB machine.",
"shouldTrigger": true,
"expect": "Applies memory discipline and model priority guidance",
"assertions": [
"Prefers quantized/MLX models over FP8",
"Uses 3-5s clips, low-res + upscale, tiling",
"Stays under ~12-13 GB peak"
]
},
{
"id": "e3",
"description": "Should trigger: character consistency across a take",
"input": "Make a 30-second talking video but keep the face consistent.",
"shouldTrigger": true,
"expect": "Uses segmented renders re-anchored on the identity sheet",
"assertions": [
"Splits at sentence boundaries into per-sentence A2V chunks",
"Each segment re-anchors on identity image, then lossless concat",
"Karaoke burn + QA after concat"
]
},
{
"id": "n1",
"description": "Not trigger: cloud rendering request",
"input": "Render this video on the cloud render farm instead.",
"shouldTrigger": false,
"expect": "Notes cloud requires explicit approval; does not run local M5 pipeline as default",
"assertions": [
"No M5 local pipeline started without approval"
]
},
{
"id": "n2",
"description": "Not trigger: primary Mac task, not M5 executor video",
"input": "Edit the audio of our podcast on the main Mac.",
"shouldTrigger": false,
"expect": "Out of scope for the M5 executor doctrine",
"assertions": [
"No MPS/LTX workflow engaged"
]
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |