Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →gooseworks-ai/render-absurdist-explainer
Assemble an absurdist animated-explainer video ad (~38s, 9:16) from per-scene i2v clips + their measured VO windows — retime each clip to its VO, re-encode every segment to identical 30fps/libx264/yuv420p so the concat demuxer never drops frames, concat, build a REAL-product PIL end card (never AI) with a slow Ken-Burns, mix VO (loudnorm I=-14) under music (loudnorm I=-26, volume 0.62, amix normalize=0), and burn libass captions last. FREE deterministic assembly (bash-free, Python + ffmpeg + PIL
Passed 22 of 22 test cases with the skill, and 10 of 22 without it.
and 17 more in the full report
12 fixed10 passed either way0 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
This benchmark measured an earlier version of the skill.
One of 12 test cases the skill fixed. Same model, same prompt, with and without the skill.
What font size and vertical margin (`MarginV`) setting should be specified in the ASS subtitle configuration for 1080x1920 video?
| Checked | Without skill | With skill |
|---|---|---|
| Sets subtitle font size to Arial 64. | Missed | Passed |
| Sets MarginV to 330. | Missed | Passed |
The instructions your model reads when the skill loads.
The free, deterministic renderer for the absurdist-explainer video ad format — an animated spot that explains one problem and how the product fixes it. The format is the mechanics: show the problem in a funny, exaggerated cartoon way, teach the product's ownable mechanism visually, show what the problem costs, bring the product in, land the fix as the climax, pay it off, end on a real-product card. One narrator voice carries the whole spot. How the story is told (problem → fix, how it works, a day in the life, the product as hero, or a villain arc) is the user's choice, made upstream. This capability is the FREE assembly stage only. All generative work (nano-banana keyframes, Seedance i2v clips, ElevenLabs VO + music) happens upstream in the recipe and is handed to this capability as files.
It ports the validated compose recipe from two reference runs (a cortisol/stress supplement and a baby-eczema cream; both happened to use the villain arc). The recipe is deterministic — iterate the cut for free, re-roll only the offending paid beat.
The creative calls are made upstream by the recipe's choices and arrive here only as
files and config values. This renderer is story- and style-agnostic: it never assumes a
story shape, a cast, a look, a narrator or a music style.
story_shape — how the story is told: problem → fix (no villain) / how it works /
a day in the life / the product as hero / villain arc (the problem as a cartoon villain
who schemes and loses). Reaches this capability as the per-scene caption text and the
VO files. Asked of the user; the demo used the villain arc.cast — who the cartoon characters are (the customer, the product as a character,
a mascot; a personified problem only on the villain arc). Reaches this capability only
inside the clips. Asked of the user.narrator — who speaks the single VO track. Arrives as the scenes[].vo files.
Asked of the user; the demo used the villain (villain arc only).visual_style — the art style of the i2v clips (scenes[].clip). Asked of the
user; the demo used Pixar-style 3D. Nothing here depends on it.narrator_voice — the voice cast for the VO. Asked of the user; the demo used a
characterful male villain voice.music — the bed at music_bed. Asked of the user; the demo used whimsical
pizzicato + woodwinds + xylophone. The mix constants below apply to any style.test cases passed, vs 10 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.