Anil-matcha/vox-ai-motion-graphics-generator
muapi-director
VerifiedTurn ONE topic, talking-head video, or photo into a finished Vox-style paper-collage explainer / ad video on the MuAPI platform (api.muapi.ai) + local ffmpeg — script, collage keyframes, motion, voice-over, music, captions, all automated. Three input modalities: a topic (B-roll), a talking-head video (A-roll mode), or a single photo of a person/product anchored into the collage (C-roll mode).
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.7 KB |
| assets/demo.mp4 | binary | 6.3 MB |
| assets/youtube-thumbnail.jpg | binary | 100.1 KB |
| examples/sample-beats.json | config | 2.4 KB |
| references/beat-layer.md | markdown | 3.3 KB |
| references/prompt-guide.md | markdown | 3.0 KB |
| references/voices.md | markdown | 1.4 KB |
| scripts/aroll_assemble.py | script | 2.6 KB |
| scripts/aroll_clips.py | script | 4.5 KB |
| scripts/asr_beats.py | script | 4.9 KB |
| scripts/assemble.py | script | 6.4 KB |
| scripts/audio.py | script | 3.7 KB |
| scripts/clips.py | script | 5.8 KB |
| scripts/croll_keyframes.py | script | 5.1 KB |
| scripts/keyframes.py | script | 3.1 KB |
| scripts/muapi_client.py | script | 9.0 KB |
| scripts/provider.py | script | 5.5 KB |
| scripts/style_bakeoff.py | script | 2.4 KB |
| scripts/styles.py | script | 11.5 KB |
| scripts/text_overlay.py | script | 4.6 KB |
scripts/aroll_clips.py
View on GitHub#!/usr/bin/env python3
"""
A-roll clip stage: cut each ASR beat's time range out of the source recording
and re-style it as a paper-collage clip via MuAPI video models.
"""
import json
import os
import subprocess
import sys
from provider import get_provider, run_jobs
from styles import resolve_theme, resolve_video_aspect
PRIMARY_MODEL = "veo31-image-to-video"
def cut_segment(src, start, end, dest, pad=0.15):
dur = max(end - start, 0.1) + pad
ss = max(start - pad / 2, 0)
subprocess.run(["ffmpeg", "-y", "-v", "error", "-ss", f"{ss:.2f}", "-i", src,
"-t", f"{dur:.2f}", "-c:v", "libx264", "-c:a", "aac",
"-pix_fmt", "yuv420p", dest], check=True)
def build_prompt(theme, content_beats=""):
tp = theme or {}
idiom = tp.get("idiom", "american-retro")
palette = tp.get("palette", "bold retro primaries")
type_style = tp.get("type_style", "bold condensed headlines")
finish = tp.get("finish", "aged paper, high contrast")
beats_line = f"\n\nCONTENT BEATS: {content_beats}" if content_beats else ""
return (
"Transform my raw talking-head video into a mixed-media collage motion graphic.\n\n"
"TALKING-HEAD LOCK (most important):\n"
"The presenter is a PHOTOGRAPHIC paper cut-out sticker with a bold black printed "
"outline and a torn-paper silhouette edge -- her actual photographic likeness, "
"facial performance, LIP MOVEMENTS, eye-line and hand gestures follow the source "
"video frame-for-frame, unmodified. Do NOT re-texture, redraw, or halftone-print the "
"face or skin itself -- only the silhouette edge and the world around her are paper-"
"collage. Keep ALL motion, timing and framing from the source.\n\n"
f"STYLE (background & props): {idiom}. Palette: {palette}. Typography: {type_style}. "
f"Finish: {finish}. Clearly layered hand-cut paper cut-outs with visible torn and "
"scissor-cut edges, tape corners and soft real paper drop shadows, on a bold flat "
f"paper background.{beats_line}\n\n"
"CONSTRAINTS: background elements and stamps never cover her face or hands."
)
def run(project_dir, only=None):
bpath = os.path.join(project_dir, "beats.json")
with open(bpath) as f:
doc = json.load(f)
if doc.get("mode") != "aroll":
raise SystemExit(f"{bpath} is not an A-roll project (mode != 'aroll')")
src = doc["source_video"]
target_aspect = doc.get("aspect", "9:16")
model = doc.get("video_model", PRIMARY_MODEL)
resolved_aspect, exact = resolve_video_aspect(target_aspect, model)
theme = resolve_theme(doc.get("theme")) or {}
prov = get_provider(doc.get("provider"))
seg_dir = os.path.join(project_dir, "_aroll_segs")
clip_dir = os.path.join(project_dir, "clips")
os.makedirs(seg_dir, exist_ok=True)
os.makedirs(clip_dir, exist_ok=True)
specs, by_id = {}, {}
for beat in doc["beats"]:
bid = str(beat["id"])
if only and bid not in only:
continue
if beat.get("clip_url") and os.path.exists(beat.get("clip_path", "")):
continue
seg_file = os.path.join(seg_dir, f"seg_{bid}.mp4")
if not os.path.exists(seg_file):
cut_segment(src, beat["start"], beat["end"], seg_file)
seg_url = prov.upload(seg_file)
beat["source_clip_url"] = seg_url
prompt = build_prompt(theme, beat.get("content_beats", ""))
beat["clip_prompt"] = prompt
params = {"image_url": seg_url, "duration": int(beat.get("dur", 5))}
if resolved_aspect:
params["aspect_ratio"] = resolved_aspect
specs[bid] = (lambda p=prompt, par=params: prov.submit_video(model, p, **par))
by_id[bid] = beat
done = run_jobs(prov, specs, poll_s=4, stall_s=120, max_retries=2, deadline_s=900)
for bid, url in done.items():
if not url:
continue
dest = os.path.join(clip_dir, f"clip_{bid}.mp4")
prov.download(url, dest)
beat = by_id[bid]
beat["clip_url"] = url
beat["clip_path"] = dest
print(f"[{bid}] saved {dest}")
with open(bpath, "w") as f:
json.dump(doc, f, ensure_ascii=False, indent=2)
print("Updated", bpath)
if __name__ == "__main__":
proj = sys.argv[1] if len(sys.argv) > 1 else os.path.join(
os.path.dirname(__file__), "..", "out", "aroll-demo")
only_ids = sys.argv[2].split(",") if len(sys.argv) > 2 else None
run(os.path.abspath(proj), only_ids)
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |