davepoon/buildwithclaude / plugins/all-skills/skills/team-headshot-consistency
team-headshot-consistency
VerifiedMake a website's team or about page photos look like one set — scan the headshot folder locally, flag photos whose aspect ratio, brightness, colour temperature, background or resolution differ from the rest, write a shared photo spec, and optionally export every photo at the same crop and size. Use when building or fixing a team page, "our team photos look inconsistent", "make all headshots the same size", "standardize staff photos", or when adding a new hire's photo to an existing page. Uses the free ProfessionalHeadshot.io MCP for framing guidance and photo scores when connected; the local scan works without it.
| File | Kind | Size |
|---|---|---|
| LICENSE | other | 1.1 KB |
| SKILL.md | skill md | 4.3 KB |
| references/report-template.md | markdown | 545 B |
| references/team-photo-spec.md | markdown | 1.3 KB |
| scripts/team_audit.py | script | 5.4 KB |
scripts/team_audit.py
View on GitHub#!/usr/bin/env python3
"""Measure a folder of team headshots and flag the ones that do not match the set.
Usage:
python3 team_audit.py DIR [--json]
python3 team_audit.py DIR --normalize OUT_DIR [--size 800] [--ratio 1:1] [--bg "#F2F2F2"]
Everything runs locally; nothing is uploaded. Needs Pillow (pip install pillow).
The audit measures pixels only (size, aspect ratio, brightness, colour temperature,
border colour); it does not detect faces, so framing must be checked by eye.
"""
from __future__ import annotations
import argparse
import json
import statistics
import sys
from pathlib import Path
try:
from PIL import Image, ImageOps, ImageStat
except ImportError: # pragma: no cover
sys.exit("Pillow is required: pip install pillow")
EXTENSIONS = {".jpg", ".jpeg", ".png", ".webp"}
# A photo is flagged when it sits this far from the team median.
RATIO_TOLERANCE = 0.03 # aspect ratio difference
BRIGHTNESS_TOLERANCE = 25 # 0-255 luma
WARMTH_TOLERANCE = 18 # red minus blue, 0-255
BORDER_TOLERANCE = 40 # mean RGB distance of the outer 6% border
MIN_SHORT_EDGE = 400
def measure(path: Path) -> dict:
img = ImageOps.exif_transpose(Image.open(path)).convert("RGB")
w, h = img.size
stat = ImageStat.Stat(img)
r, g, b = stat.mean
luma = 0.299 * r + 0.587 * g + 0.114 * b
edge = max(1, round(min(w, h) * 0.06))
strips = [img.crop((0, 0, w, edge)), img.crop((0, 0, edge, h)), img.crop((w - edge, 0, w, h))]
border = [statistics.fmean(ImageStat.Stat(s).mean[i] for s in strips) for i in range(3)]
return {
"file": path.name, "width": w, "height": h, "ratio": round(w / h, 3),
"brightness": round(luma, 1), "warmth": round(r - b, 1),
"border": [round(c) for c in border],
}
def distance(a: list[float], b: list[float]) -> float:
return sum((x - y) ** 2 for x, y in zip(a, b)) ** 0.5
def audit(rows: list[dict]) -> dict:
med = {
"ratio": statistics.median(r["ratio"] for r in rows),
"brightness": statistics.median(r["brightness"] for r in rows),
"warmth": statistics.median(r["warmth"] for r in rows),
"border": [statistics.median(r["border"][i] for r in rows) for i in range(3)],
}
for r in rows:
flags = []
if min(r["width"], r["height"]) < MIN_SHORT_EDGE:
flags.append(f"low resolution (short edge {min(r['width'], r['height'])} px)")
if abs(r["ratio"] - med["ratio"]) > RATIO_TOLERANCE:
flags.append(f"aspect ratio {r['ratio']} vs team {med['ratio']}")
if abs(r["brightness"] - med["brightness"]) > BRIGHTNESS_TOLERANCE:
flags.append("much darker than the team" if r["brightness"] < med["brightness"] else "much brighter than the team")
if abs(r["warmth"] - med["warmth"]) > WARMTH_TOLERANCE:
flags.append("warmer colour than the team" if r["warmth"] > med["warmth"] else "cooler colour than the team")
if distance(r["border"], med["border"]) > BORDER_TOLERANCE:
flags.append("background colour differs from the team")
r["flags"] = flags
return {"teamMedian": med, "photos": rows, "outliers": [r["file"] for r in rows if r["flags"]]}
def normalize(paths: list[Path], out: Path, size: int, ratio: float, bg: str) -> list[str]:
"""Same crop ratio and output size for everyone; pads (never stretches) when needed."""
out.mkdir(parents=True, exist_ok=True)
w, h = (size, round(size / ratio)) if ratio >= 1 else (round(size * ratio), size)
written = []
for p in paths:
img = ImageOps.exif_transpose(Image.open(p)).convert("RGB")
fitted = ImageOps.fit(img, (w, h), Image.LANCZOS, centering=(0.5, 0.4))
canvas = Image.new("RGB", (w, h), bg)
canvas.paste(fitted, (0, 0))
dest = out / f"{p.stem}.jpg"
canvas.save(dest, "JPEG", quality=90, optimize=True)
written.append(str(dest))
return written
def parse_ratio(raw: str) -> float:
a, _, b = raw.partition(":")
try:
value = float(a) / float(b)
except (ValueError, ZeroDivisionError) as exc:
raise argparse.ArgumentTypeError("use W:H, e.g. 1:1 or 4:5") from exc
return value
def main(argv: list[str] | None = None) -> int:
parser = argparse.ArgumentParser(description=__doc__.splitlines()[0])
parser.add_argument("dir", type=Path)
parser.add_argument("--json", action="store_true")
parser.add_argument("--normalize", type=Path, metavar="OUT_DIR")
parser.add_argument("--size", type=int, default=800)
parser.add_argument("--ratio", type=parse_ratio, default=1.0)
parser.add_argument("--bg", default="#F2F2F2")
args = parser.parse_args(argv)
paths = sorted(p for p in args.dir.iterdir() if p.suffix.lower() in EXTENSIONS) if args.dir.is_dir() else []
if len(paths) < 2:
parser.error("need a folder with at least two .jpg/.png/.webp photos")
if args.normalize:
print(json.dumps({"written": normalize(paths, args.normalize, args.size, args.ratio, args.bg)}, indent=2))
return 0
report = audit([measure(p) for p in paths])
if args.json:
print(json.dumps(report, indent=2))
else:
for r in report["photos"]:
status = "OK " if not r["flags"] else "FIX"
print(f"{status} {r['file']}: {'; '.join(r['flags']) or 'matches the set'}")
print(f"\n{len(report['outliers'])} of {len(paths)} photos differ from the team median.")
return 0
if __name__ == "__main__":
raise SystemExit(main())
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |