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/provider.py
View on GitHub#!/usr/bin/env python3
"""
Provider abstraction — pluggable media backend for Vox Director pipeline stages.
Supports 'muapi' as the primary provider (via MuAPI platform) and 'atlas_cloud' as fallback.
"""
import time
from abc import ABC, abstractmethod
import muapi_client
class ProviderError(RuntimeError):
pass
class Provider(ABC):
"""Abstract provider surface for Vox Director pipeline stages."""
name = "base"
@abstractmethod
def submit_image(self, model, prompt, **params): ...
@abstractmethod
def submit_video(self, model, prompt, **params): ...
@abstractmethod
def submit_audio(self, model, **params): ...
@abstractmethod
def remove_bg(self, model, image_url, **params): ...
@abstractmethod
def get_status(self, job_id): ...
@abstractmethod
def upload(self, path): ...
@abstractmethod
def download(self, url, dest): ...
class MuapiProvider(Provider):
"""Wraps muapi_client — interfaces with api.muapi.ai"""
name = "muapi"
def submit_image(self, model, prompt, **params):
return muapi_client.submit(model, {"prompt": prompt, **params})
def submit_video(self, model, prompt, **params):
return muapi_client.submit(model, {"prompt": prompt, **params})
def submit_audio(self, model, **params):
return muapi_client.submit(model, params)
def remove_bg(self, model, image_url, **params):
return muapi_client.submit(model or "remove-background", {"image_url": image_url, **params})
def get_status(self, job_id):
res = None
try:
res = muapi_client._get(f"/predictions/{job_id}/result")
except Exception as e:
try:
res = muapi_client._get(f"/predictions/{job_id}")
except Exception as e2:
return {"status": "pending", "output": None, "error": str(e2)}
if not res:
return {"status": "pending", "output": None, "error": None}
status = (res.get("status") or (res.get("data") or {}).get("status") or "").lower()
if status in ("completed", "success", "succeeded"):
url = muapi_client.extract_url(res)
return {"status": "completed", "output": url, "error": None}
if status in ("failed", "cancelled"):
err = res.get("error") or (res.get("data") or {}).get("error") or "Job failed"
return {"status": "failed", "output": None, "error": str(err)}
return {"status": "pending", "output": None, "error": None}
def upload(self, path):
return muapi_client.upload(path)
def download(self, url, dest):
return muapi_client.download(url, dest)
_REGISTRY = {
"muapi": MuapiProvider,
}
def get_provider(name=None):
"""Return a Provider instance by name (default 'muapi')."""
name = (name or "muapi").lower()
if name not in _REGISTRY:
raise ProviderError(f"Unknown provider '{name}'; available: {list(_REGISTRY)}")
return _REGISTRY[name]()
def run_jobs(prov, specs, *, poll_s=3, stall_s=90, max_retries=2, deadline_s=900):
"""Submit + poll a batch of jobs with resubmission on failure or stall."""
state = {}
for key, submit_fn in specs.items():
try:
jid = submit_fn()
state[key] = {
"submit_fn": submit_fn,
"id": jid,
"tries": 1,
"last_change": time.monotonic(),
"out": None,
"error": None,
"done": False,
}
print(f" [{key}] submitted -> {jid}")
except Exception as e:
print(f" [{key}] submit failed: {e}")
state[key] = {
"submit_fn": submit_fn,
"id": None,
"tries": 1,
"last_change": time.monotonic(),
"out": None,
"error": str(e),
"done": True,
}
deadline = time.monotonic() + deadline_s
while time.monotonic() < deadline:
all_done = True
for key, s in state.items():
if s["done"]:
continue
all_done = False
res = prov.get_status(s["id"])
st = res["status"]
if st == "completed":
s["out"] = res["output"]
s["done"] = True
print(f" [{key}] completed -> {s['out']}")
elif st == "failed" or (time.monotonic() - s["last_change"] > stall_s):
reason = res["error"] if st == "failed" else f"stalled >{stall_s}s"
if s["tries"] <= max_retries:
print(f" [{key}] {reason}; retry {s['tries']}/{max_retries}...")
try:
new_id = s["submit_fn"]()
s["id"] = new_id
s["tries"] += 1
s["last_change"] = time.monotonic()
print(f" [{key}] re-submitted -> {new_id}")
except Exception as e:
s["error"] = str(e)
s["done"] = True
print(f" [{key}] re-submit failed: {e}")
else:
s["error"] = reason
s["done"] = True
print(f" [{key}] failed permanently: {reason}")
if all_done:
break
time.sleep(poll_s)
results = {}
for key, s in state.items():
results[key] = s["out"]
if not s["out"]:
print(f" WARNING: [{key}] produced no output (error: {s['error']})")
return results
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 |