Skip to content

Open-source AI podcast clipper. Generate vertical clips with face tracking and burned-in captions. CLI, MCP server, web app.

feat: default quality max, auto-detect outro from registered assets

Nika Siradze5fdcf79530c9
diff --git a/backend/cli.py b/backend/cli.py
index 632106c..3269bd9 100644
--- a/backend/cli.py
+++ b/backend/cli.py
@@ -157,6 +157,16 @@ def cmd_process(args):
else:
print(f" Warning: Outro '{args.outro}' not found (checked assets and filesystem)", file=sys.stderr)
config["outro_path"] = args.outro
+ elif not config.get("outro_path"):
+ # Auto-detect outro from registered assets
+ try:
+ from services.asset_store import list_assets as _list_assets_auto
+ for a in _list_assets_auto():
+ if a["type"] == "video" and os.path.exists(a["path"]):
+ config["outro_path"] = a["path"]
+ break
+ except Exception:
+ pass
if args.time_adjust is not None:
config["time_adjust"] = args.time_adjust
if args.no_energy:
@@ -169,7 +179,7 @@ def cmd_process(args):
config["quality"] = args.quality
# Set quality env var before importing video_processor
- quality = config.get("quality", os.environ.get("PODCLI_QUALITY", "high"))
+ quality = config.get("quality", os.environ.get("PODCLI_QUALITY", "max"))
os.environ["PODCLI_QUALITY"] = quality
# Output directory