gooseworks-ai/goose-skills / skills/video/render-brand-identity-reveal
render-brand-identity-reveal
VerifiedRender a 'brand identity reveal' video from a config — a single poster frame in a real, softly-lit space (real wall, soft-focus plant in the corner, dappled leaf shadow, illuminated poster) whose artwork HARD-CUTS through ~10 on-brand poster mockups (hero product, IG post, hanging banners, sticker sheet, logo lockup, poster, two lifestyle stills, packaging, big icon) then holds on a brand end card. The environment plate is one create-image-fal generation; the mockups are real-DOM HTML frame-stepped via Playwright, perspective-composited into the detected frame quad with the plate's real leaf-shadow multiplied back onto each poster (reads as behind glass), sequenced by FFmpeg. Deterministic assembly, FREE (the plate comes from create-image-fal, the bed from create-music-elevenlabs), music bed only and approved brand copy only. Use for the brand-identity-reveal format.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 5.3 KB |
| scripts/PIPELINE.md | markdown | 3.5 KB |
| scripts/build_video.shexecutable | script | 1.2 KB |
| scripts/composite.py | script | 3.0 KB |
| scripts/config.example.json | config | 3.9 KB |
| scripts/measure_frame.py | script | 1.7 KB |
| scripts/recrop.py | script | 2.2 KB |
| scripts/render_art.py | script | 1.3 KB |
| scripts/scene.html | other | 20.4 KB |
| skill.meta.json | config | 223 B |
| tests/smoke-test.md | markdown | 1.1 KB |
scripts/composite.py
View on GitHub#!/usr/bin/env python3
"""Composite each standalone artwork into the lit poster frame of the photoreal plate.
Reintroduces the plate's real leaf-shadow/lighting onto the poster + adds glass sheen."""
import os, ast
import numpy as np
from PIL import Image, ImageDraw, ImageFilter
HERE = os.path.dirname(os.path.abspath(__file__))
PLATE = Image.open(os.path.join(HERE, "bg/plate_final.png")).convert("RGB")
W, H = PLATE.size
c = ast.literal_eval(open(os.path.join(HERE, "bg/corners.txt")).read())
# inset 4px so artwork never bleeds onto the wood frame
def inset(p, dx, dy): return (p[0] + dx, p[1] + dy)
TL = inset(c["tl"], 4, 4); TR = inset(c["tr"], -4, 4)
BR = inset(c["br"], -4, -4); BL = inset(c["bl"], 4, -4)
QUAD = [TL, TR, BR, BL]
OUT = os.path.join(HERE, "frames_v2"); os.makedirs(OUT, exist_ok=True)
def find_coeffs(output_pts, input_pts):
A = []
for (xo, yo), (xi, yi) in zip(output_pts, input_pts):
A.append([xo, yo, 1, 0, 0, 0, -xi * xo, -xi * yo])
A.append([0, 0, 0, xo, yo, 1, -yi * xo, -yi * yo])
A = np.array(A, dtype=np.float64)
B = np.array(input_pts, dtype=np.float64).reshape(8)
return np.linalg.solve(A, B)
# --- shadow/light map from the empty plate interior (canvas space) ---
plate_np = np.asarray(PLATE).astype(np.float32)
luma = 0.2126 * plate_np[..., 0] + 0.7152 * plate_np[..., 1] + 0.0722 * plate_np[..., 2]
# reference "fully-lit" white = high percentile inside the quad bbox
xs = [p[0] for p in QUAD]; ys = [p[1] for p in QUAD]
sub = luma[min(ys):max(ys), min(xs):max(xs)]
ref = np.percentile(sub, 97)
mapf = np.clip(luma / max(ref, 1.0), 0.25, 1.0)
mapf = 1.0 - (1.0 - mapf) * 0.85 # 0.85 = shadow strength
mapf = np.stack([mapf] * 3, axis=-1) # HxWx3
# glass sheen (soft diagonal highlight) + edge inner shadow, built once in quad space
sheen = Image.new("L", (W, H), 0)
sd = ImageDraw.Draw(sheen)
sd.polygon(QUAD, fill=0)
# a bright band across the upper-left of the poster
band = Image.new("L", (W, H), 0)
bd = ImageDraw.Draw(band)
bx0, by0 = TL; bx1 = TR[0]
bd.polygon([(bx0, by0), (bx0 + (bx1 - bx0) * 0.55, by0),
(bx0, by0 + (BL[1] - by0) * 0.62)], fill=42)
band = band.filter(ImageFilter.GaussianBlur(70))
sheen_np = np.asarray(band).astype(np.float32)[..., None]
def coeffs_for():
inp = [(0, 0), (1480 - 1, 0), (1480 - 1, 2136 - 1), (0, 2136 - 1)]
return find_coeffs(QUAD, inp)
COEFFS = coeffs_for()
for i in range(1, 12):
art = Image.open(os.path.join(HERE, f"art/art_std_{i:02d}.png")).convert("RGBA")
warped = art.transform((W, H), Image.PERSPECTIVE, COEFFS, Image.BICUBIC, fillcolor=(0, 0, 0, 0))
wnp = np.asarray(warped).astype(np.float32)
rgb = wnp[..., :3]; alpha = wnp[..., 3:4] / 255.0
rgb = rgb * mapf # bake real leaf-shadow onto poster
rgb = np.clip(rgb + sheen_np, 0, 255) # glass sheen highlight
out = plate_np.copy()
out = out * (1 - alpha) + rgb * alpha # alpha composite over plate
Image.fromarray(out.astype(np.uint8)).save(os.path.join(OUT, f"frame_{i:02d}.png"))
print("wrote", f"frame_{i:02d}.png")
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |