一键在 Manus 中运行任何 Skill

render-creator-pip-listicle

星标1,046

分支193

更新时间2026年7月19日 03:10

Assemble a creator picture-in-picture product-listicle ad from a config — the creator stays FULL-FRAME the whole beat (voice plus lips generated together per beat, no separate VO, no cut to a full-frame product shot), and on each product beat three persistent overlays ride on top for the WHOLE beat — a title pill top-center, the DEMO in a rounded PiP window top-right (the brand's real UGC clip MUTED, or for a no-UGC brand the product's own autocropped UI still / screen-recording sized to fill the window), and a bottom product card (rounded thumbnail plus 'N · CATEGORY' small-caps plus product NAME in a serif face). Hook plus CTA beats are the creator full-frame with the title pill only. Assembly builds ONE full-1080x1920 transparent overlay PNG per beat, overlays it on the creator clip (cover-scaled to 1080x1920) for the whole beat keeping the native audio, concats all beats, then burns captions LAST as timed PIL PNG overlays (this ffmpeg has no libass) timed deterministically from the known per-beat script.

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

gooseworks-ai

gooseworks-ai/goose-skills

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

render-creator-pip-listicle

Assemble a creator picture-in-picture product listicle ad from a config: an AI creator counts down N products in the brand's own voice, and the creator stays FULL-FRAME the whole time — there is NO cut to a full-frame product shot, ever. On each product beat, three persistent overlays ride on top of the full-frame creator for the WHOLE beat: (1) a title pill top-center (persistent, it carries the listicle title), (2) the DEMO in a rounded PiP window top-right — the brand's real UGC clip (MUTED), or for a no-UGC brand the product's own demo (a real screen-recording, or an autocropped high-res product-UI/dashboard still sized to fill the window), and (3) a product card pinned bottom (rounded thumbnail left + "N · CATEGORY" small-caps + product NAME in a serif face like Georgia). Hook + CTA beats are the creator full-frame with the title pill only (no PiP/card). The creator's voice + lips are generated together, natively per beat — there is no separate voiceover. This capability is the FREE, deterministic assembly — build the per-beat overlay PNG, cover-scale the creator clip + composite the overlay, concat all beats, and burn the captions.

scripts/config.example.json is the worked example (DIBS Beauty "5 products that replaced my whole makeup bag", ~46s 1080×1920 9:16, a hook + 5 product beats + a CTA); scripts/PIPELINE.md maps every config block to its source step and scripts/README.md documents the free assembly.

Run

This is the FREE, deterministic assembly stage — it spends nothing beyond the caption burn. The paid inputs are separate capabilities — the creator anchor (create-image-fal, Seedream v5 Pro, model bytedance/seedream/v5/pro/text-to-image with no fal-ai/ prefix) and one native Seedance talking clip per beat (create-video-fal, model bytedance/seedance-2.0/reference-to-video, generate_audio=ON, the SAME seed across beats so the face holds, 720p default). Given those native clips + the brand's real UGC demo clips (or the product's own autocropped UI stills / screen recordings) + the real product photos + the brand palette + the title copy, render-creator-pip-listicle builds ONE full-1080×1920 transparent overlay PNG per beat (title pill always; + demo PiP top-right + bottom product card + rank number on product beats), cover-scales the creator clip to 1080×1920 and overlays the beat's overlay PNG for the whole beat while keeping the native audio, concats all beats, and burns the captions last → the master. Re-cuts reuse the existing native clips + overlays and cost $0.

Anchor gotcha (Seedream v5 Pro, not gpt-image-2/Seedream-v4). Seedance 2.0's partner-validation gate REJECTS photoreal faces from gpt-image-2 AND Seedream v4 ("may contain likenesses of real people"); a Seedream v5 Pro face passes. Generate a FRESH anchor with create-image-fal, model bytedance/seedream/v5/pro/text-to-image (no fal-ai/ prefix) — reusing an existing photoreal face from another project also trips the gate.

Clip gotchas (Seedance 2.0). A REJECTED submit STILL bills via the fal-proxy — pre-flight ONE test clip (the hook) and confirm it renders before firing the batch. Presigned anchor URLs expire ~1h → re-host the anchor if a batch runs long (else "Failed to download the file" mid-batch). The intermittent fal "User is locked: Exhausted balance" is the proxy's upstream fal account (not your GooseWorks credits) → retry with backoff.

Contract (the free assembly)

Creator FULL-FRAME the whole beat — NO cut to a full-frame product shot. Every beat is ONE continuous full-frame creator clip; the product content lives in overlays, never in a full-frame cutaway. per_beat_shots = 1.
Native creator audio carries the reel — no separate VO. Each beat's voice + lips come from ONE Seedance take (generate_audio=ON); this stage never adds a VO or a lip-sync pass. The creator clip's native audio plays continuous across the whole beat; the demo PiP's audio is muted (else the voice doubles).
Three persistent overlays on each product beat. For the WHOLE beat, on top of the full-frame creator: (a) the title pill top-center (persistent, title_pill.on=true — carries the listicle title, kept short to fit one line); (b) the DEMO PiP top-right; (c) the product card bottom. Hook + CTA carry the title pill only.
The DEMO PiP is autocropped to FILL its window. It's the brand's REAL UGC clip (MUTED, never AI-regenerated) — OR, for a brand with NO UGC (B2B/SaaS), the product's own demo: a real screen-recording, or an AUTOCROPPED high-res product-UI/dashboard still. Autocrop the still (trim transparent/near-white margins) and size the PiP window to the cropped content's aspect ratio so it fills the window — no letterbox whitespace; a WIDE screenshot → a SHORT + WIDE window, a tall/square one → a taller window. Rounded window, white hairline border, soft drop shadow, top-right. Disclose in the review when the demo is a still/mockup rather than a real UGC clip.
Products are REAL photos — never AI-render the product. The bottom product card (rounded thumbnail + "N · CATEGORY" small-caps + product NAME in a serif face like Georgia) comes from the brand's real product photo / UI thumbnail; a product with no clean photo falls back to a brand-color tile with the wordmark, never an AI product render.
Persistent title pill, rank number on the card. The title pill is persistent top-center (this format's identity, on by default). A counting rank number (1..N) is rendered on the product card on product beats.
Captions burned LAST as PIL PNG overlays, brand-accent, deterministic. This ffmpeg has no libass → render each ~2-word cue as a timed PIL PNG overlay (overlay=…:enable='between(t,s,e)'): white words + a brand-accent underline, black stroke for legibility, positioned CLEAR of the PiP (top) and the card (bottom) (mid-to-lower band). Time them DETERMINISTICALLY from the known per-beat script — the fal-ai/whisper proxy is unreliable (900s timeouts); do NOT depend on it. The known script is the brand-correct source, so brand tokens are always spelled right.
FFmpeg composite, deterministic, FREE. Per beat: build ONE full-1080×1920 transparent overlay PNG (title pill; + PiP + card + rank number on product beats), cover-scale the creator clip to 1080×1920, overlay the PNG for the whole beat, keep the native audio. Concat all beats with the concat demuxer @ 30fps / yuv420p → a 1080×1920 h264+aac master, then burn captions last. Probe durations with ffprobe -of csv=p=0 (NOT -of default=nk=1:np=1, which errors on some builds). No paid calls in the composite/stitch, no keys.

同仓库更多 Skills

同仓库

create-image-gpt-image-fal

gooseworks-ai/goose-skills

Generate a single photoreal or designed image with OpenAI gpt-image via fal.ai. Supports gpt-image-1 (default, fixed sizes — the FAL fallback for Higgsfield's `gpt_image_2`) and gpt-image-2 (`openai/gpt-image-2`, custom output sizes up to 3840px). Routes to text-to-image or the edit variant depending on whether a reference image is provided. Use for photoreal character anchors, scene keyframes, and designed sheets (e.g. storyboards) where precise layout and legible text matter.

2026-07-191.0k

render-cartoon-music-video

gooseworks-ai/goose-skills

Assemble a cartoon / animated / hand-crafted music-video ad from a config — a sung song carries the whole narrative while N per-bar i2v clips (one recurring animated character, one look pack) are each cut to their BAR window from librosa beat-tracking and hard-concatenated on the bar, VEED-whisper white bold-sans captions in the BOTTOM third (Alignment 2, above the logo bug, no pill) burned from the song's word timings re-spelled against the locked lyrics, a persistent brand logo bug held over the body (suppressed on the end card), and closed on a solid-brand-color PIL end card with the song still playing under it — never AI-rendered text. This is the FREE deterministic assembly stage (cut-to-bar + hard concat + logo bug + captions + end card + song mux); the song, character, keyframes, and clips come from create-music-elevenlabs / create-image-fal / create-video-fal. Use for the cartoon-music-video format.

2026-07-191.0k

render-cosmic-mythology-voiceover

gooseworks-ai/goose-skills

Assemble a cosmic-mythology-voiceover reel from a config — a warm spoken voiceover carries the whole narrative while N curated cosmic stills are weighted beat-synced across the delivered VO duration (cut_dur = VO_dur times weight over the weight sum, so emotional beats hold longer), Ken-Burns-zoomed per still (scale 2x, center crop, zoompan, fade-in first and fade-out last), ffmpeg-concatenated, the VO composited under the picture (libx264 crf18 plus aac), the ONE on-screen hook line faded on over the open with a drawtext alpha window, and Whisper/VEED captions burned along the bottom — never in-world text on a still. This is the FREE deterministic assembly stage (weighted sequence plus Ken-Burns plus concat plus VO composite plus hook overlay plus caption burn); the VO and the stills come from create-vo-elevenlabs and create-image-fal. Use for the cosmic-mythology-voiceover format.

2026-07-191.0k

render-food-product-sizzle

gooseworks-ai/goose-skills

Assemble a wordless macro-tabletop food-product sizzle ad from a config — normalize fps and SAR across ~4 photorealistic macro clips (hands tearing, flat lay, bite, box hero), concat them, apply a global anti-AI grain pass (eq plus hqdn3d plus noise), composite the audio (a non-diegetic acoustic music bed plus a couple of short diegetic SFX like a snap and a tear placed at measured cue points, loudnorm), composite a STATIC end card entirely in PIL (real logo PNG plus real product PNG plus a serif heritage headline plus a CTA — never AI-rendered text), and burn optional serif stat-callout pills at beats. This is the FREE deterministic assembly stage (normalized concat plus grain plus music and SFX mix plus PIL end card plus callouts); the macro keyframes, i2v clips, and music bed come from create-image-fal, create-video-fal, and create-music-elevenlabs. Use for the food-product-sizzle format.

2026-07-191.0k

render-split-screen-creator

gooseworks-ai/goose-skills

Assemble a split-screen creator ad from a config — a two-zone vertical composite where a supplied AI-creator lip-sync take fills the BOTTOM ~48% while real 16:9 product/demo clips run uncropped in the TOP ~52%, each top clip contain-fit with a darkened blurred cover-scale fill of the same clip (never black bars), a 3px brand-color divider between the zones, the creator slice cover-fit per the per-scene VO timing, scenes hard-concatenated with the body audio being the concatenated creator VO slices, an end card held on the last sharp frame ~3s, then the ASSEMBLED cut transcribed with local Whisper (not the raw VO — concat drops inter-scene silence) and word-level captions burned in the chosen style. This is the FREE deterministic assembly + caption stage (two-zone composite + blurred fill + divider + hard-concat + end card + captions); the VO comes from create-vo-elevenlabs, the anchor from create-image-gpt-image-fal, and the whole-VO lip-sync from a paid VEED Fabric 1.0 take (a no-atom upstream input). Use fo

2026-07-191.0k

create-video-seedance-2-fal

gooseworks-ai/goose-skills

Generate a single 4-15s vertical video clip with ByteDance Seedance 2.0 reference-to-video via fal.ai. Multi-image reference (avatar + product + setting), native lip-synced VO + ambient audio (generate-audio on by default), internal multi-cut handling within one render. Routes through the GooseWorks FAL proxy (bills the Ads agent). The default clip atom for AI-creator UGC ads built on the NB2 + Seedance architecture. Validated on beauty-by-earth/video-01.

2026-07-191.0k

name	render-creator-pip-listicle
description	Assemble a creator picture-in-picture product-listicle ad from a config — the creator stays FULL-FRAME the whole beat (voice plus lips generated together per beat, no separate VO, no cut to a full-frame product shot), and on each product beat three persistent overlays ride on top for the WHOLE beat — a title pill top-center, the DEMO in a rounded PiP window top-right (the brand's real UGC clip MUTED, or for a no-UGC brand the product's own autocropped UI still / screen-recording sized to fill the window), and a bottom product card (rounded thumbnail plus 'N · CATEGORY' small-caps plus product NAME in a serif face). Hook plus CTA beats are the creator full-frame with the title pill only. Assembly builds ONE full-1080x1920 transparent overlay PNG per beat, overlays it on the creator clip (cover-scaled to 1080x1920) for the whole beat keeping the native audio, concats all beats, then burns captions LAST as timed PIL PNG overlays (this ffmpeg has no libass) timed deterministically from the known per-beat script. This is the FREE deterministic assembly stage (overlay-PNG build plus cover-scale composite plus concat plus PIL-PNG caption burn); the creator anchor and the N native talking clips come from create-image-fal (Seedream v5 Pro) and create-video-fal (Seedance 2.0). Use for the creator-pip-listicle format.
status	active

render-creator-pip-listicle

Run

Contract (the free assembly)

Creator FULL-FRAME the whole beat — NO cut to a full-frame product shot. Every beat is ONE continuous full-frame creator clip; the product content lives in overlays, never in a full-frame cutaway. per_beat_shots = 1.
Native creator audio carries the reel — no separate VO. Each beat's voice + lips come from ONE Seedance take (generate_audio=ON); this stage never adds a VO or a lip-sync pass. The creator clip's native audio plays continuous across the whole beat; the demo PiP's audio is muted (else the voice doubles).
Three persistent overlays on each product beat. For the WHOLE beat, on top of the full-frame creator: (a) the title pill top-center (persistent, title_pill.on=true — carries the listicle title, kept short to fit one line); (b) the DEMO PiP top-right; (c) the product card bottom. Hook + CTA carry the title pill only.
The DEMO PiP is autocropped to FILL its window. It's the brand's REAL UGC clip (MUTED, never AI-regenerated) — OR, for a brand with NO UGC (B2B/SaaS), the product's own demo: a real screen-recording, or an AUTOCROPPED high-res product-UI/dashboard still. Autocrop the still (trim transparent/near-white margins) and size the PiP window to the cropped content's aspect ratio so it fills the window — no letterbox whitespace; a WIDE screenshot → a SHORT + WIDE window, a tall/square one → a taller window. Rounded window, white hairline border, soft drop shadow, top-right. Disclose in the review when the demo is a still/mockup rather than a real UGC clip.
Products are REAL photos — never AI-render the product. The bottom product card (rounded thumbnail + "N · CATEGORY" small-caps + product NAME in a serif face like Georgia) comes from the brand's real product photo / UI thumbnail; a product with no clean photo falls back to a brand-color tile with the wordmark, never an AI product render.
Persistent title pill, rank number on the card. The title pill is persistent top-center (this format's identity, on by default). A counting rank number (1..N) is rendered on the product card on product beats.
Captions burned LAST as PIL PNG overlays, brand-accent, deterministic. This ffmpeg has no libass → render each ~2-word cue as a timed PIL PNG overlay (overlay=…:enable='between(t,s,e)'): white words + a brand-accent underline, black stroke for legibility, positioned CLEAR of the PiP (top) and the card (bottom) (mid-to-lower band). Time them DETERMINISTICALLY from the known per-beat script — the fal-ai/whisper proxy is unreliable (900s timeouts); do NOT depend on it. The known script is the brand-correct source, so brand tokens are always spelled right.
FFmpeg composite, deterministic, FREE. Per beat: build ONE full-1080×1920 transparent overlay PNG (title pill; + PiP + card + rank number on product beats), cover-scale the creator clip to 1080×1920, overlay the PNG for the whole beat, keep the native audio. Concat all beats with the concat demuxer @ 30fps / yuv420p → a 1080×1920 h264+aac master, then burn captions last. Probe durations with ffprobe -of csv=p=0 (NOT -of default=nk=1:np=1, which errors on some builds). No paid calls in the composite/stitch, no keys.