skilldrop-cli 0.13.4 → 0.13.6

package/packs/design/skills/marketing-flyer/scripts/build_flyer.py

@@ -0,0 +1,323 @@
+#!/usr/bin/env python3
+"""Build a print-ready, on-brand flyer as one self-contained HTML file. Stdlib only.
+
+The spec (see templates/flyer-spec.json) gives the content; a brand.json (the brand-kit
+skill writes one) gives the logo, colours and fonts. Images are embedded as data URIs, so the
+.html file is the whole flyer: open it and print, or save as PDF, at 100% scale.
+
+Usage:
+  python3 build_flyer.py spec.json -o flyer.html [--brand brand.json] [--pdf flyer.pdf] [--png flyer.png]
+
+--pdf and --png use a local Chrome or Chromium in headless mode when one is installed; without
+one the script says so and the HTML is still the deliverable.
+"""
+import argparse
+import base64
+import html
+import json
+import mimetypes
+import os
+import re
+import shutil
+import subprocess
+import sys
+from pathlib import Path
+
+SIZES = {  # (width, height) in CSS units, portrait
+    "letter": ("8.5in", "11in"), "a4": ("210mm", "297mm"), "a5": ("148mm", "210mm"),
+    "square": ("1080px", "1080px"), "story": ("1080px", "1920px"),
+}
+PX = {"letter": (816, 1056), "a4": (794, 1123), "a5": (559, 794), "square": (1080, 1080), "story": (1080, 1920)}
+LAYOUTS = ("hero", "split", "event", "minimal")
+DEFAULTS = {"primary": "#1A2A6C", "secondary": "#4A5A8C", "accent": "#FDBB2D", "background": "#FFFFFF", "text": "#222222"}
+WARNINGS = []
+
+
+def warn(msg):
+    WARNINGS.append(msg)
+    print(f"WARN {msg}", file=sys.stderr)
+
+
+def die(msg):
+    print(f"ERROR {msg}", file=sys.stderr)
+    sys.exit(1)
+
+
+def lum(hex_str):
+    h = hex_str.lstrip("#")
+    def ch(v):
+        v = int(v, 16) / 255
+        return v / 12.92 if v <= 0.03928 else ((v + 0.055) / 1.055) ** 2.4
+    r, g, b = (ch(h[i:i + 2]) for i in (0, 2, 4))
+    return 0.2126 * r + 0.7152 * g + 0.0722 * b
+
+
+def contrast(a, b):
+    la, lb = sorted((lum(a), lum(b)), reverse=True)
+    return (la + 0.05) / (lb + 0.05)
+
+
+def ink_on(bg, text, light="#FFFFFF"):
+    """The readable text colour on a filled band: white or the brand text colour, whichever
+    contrasts more — so a pale accent gets dark text instead of illegible white."""
+    return light if contrast(light, bg) >= contrast(text, bg) else text
+
+
+def data_uri(path, label):
+    if not path:
+        return None
+    p = Path(path)
+    if not p.exists():
+        warn(f"{label} not found: {p} — the flyer is built without it")
+        return None
+    mime = mimetypes.guess_type(p.name)[0] or "application/octet-stream"
+    if not mime.startswith("image/"):
+        warn(f"{label} {p.name} is not an image ({mime})")
+        return None
+    return f"data:{mime};base64,{base64.b64encode(p.read_bytes()).decode()}"
+
+
+def load_json(path):
+    try:
+        return json.loads(Path(path).read_text())
+    except (OSError, json.JSONDecodeError) as e:
+        die(f"cannot read {path}: {e}")
+
+
+def resolve(base, ref):
+    return None if not ref else (Path(ref) if Path(ref).is_absolute() else (base / ref))
+
+
+def font_stack(f, generic):
+    if not f:
+        return generic
+    fam = f.get("family") if isinstance(f, dict) else f
+    fb = (f.get("fallback") if isinstance(f, dict) else None) or generic
+    return f"'{fam}', {fb}" if fam else fb
+
+
+def esc(s):
+    return html.escape(str(s or ""), quote=True)
+
+
+def build(spec_path, out_path, brand_ref=None):
+    spec_path = Path(spec_path).resolve()
+    base = spec_path.parent
+    spec = load_json(spec_path)
+
+    brand, bbase = {}, base
+    ref = brand_ref or spec.get("brand")
+    if isinstance(ref, dict):
+        brand = ref
+    elif ref:
+        bp = resolve(base, ref) if not brand_ref else Path(ref).resolve()
+        if bp.exists():
+            brand, bbase = load_json(bp), bp.parent
+        else:
+            warn(f"brand file not found: {bp} — using default colours and no logo")
+    if not brand:
+        warn("no brand supplied — the flyer uses default colours. Run brand-kit, or pass --brand")
+
+    c = {**DEFAULTS, **{k: v for k, v in (brand.get("colors") or {}).items() if k in DEFAULTS and v},
+         **{k: v for k, v in (spec.get("colors") or {}).items() if k in DEFAULTS and v}}
+    for k, v in c.items():
+        if not re.match(r"^#[0-9A-Fa-f]{6}$", v):
+            die(f"colour {k} is {v!r}, not a 6-digit hex code")
+    if contrast(c["text"], c["background"]) < 4.5:
+        die(f"text on background contrast is {contrast(c['text'], c['background']):.2f}:1, below 4.5:1 — fix the brand colours")
+    on_primary, on_accent = ink_on(c["primary"], c["text"]), ink_on(c["accent"], c["text"])
+    fonts = brand.get("fonts") or {}
+    head_font, body_font = font_stack(fonts.get("heading"), "Georgia, serif"), font_stack(fonts.get("body"), "Helvetica, Arial, sans-serif")
+
+    logo = brand.get("logo") or {}
+    if isinstance(logo, str):
+        logo = {"primary": logo}
+    logo_light = data_uri(resolve(bbase, logo.get("primary")), "logo")
+    logo_dark = data_uri(resolve(bbase, logo.get("on_dark")), "on-dark logo") or logo_light
+    if not logo_light and not logo_dark:
+        warn("no logo — the brand name is set as text instead")
+    name = brand.get("name") or spec.get("name") or ""
+
+    size = str(spec.get("size", "letter")).lower()
+    if size not in SIZES:
+        warn(f"unknown size {size!r}; using letter")
+        size = "letter"
+    w, h = SIZES[size]
+    landscape = str(spec.get("orientation", "portrait")).lower() == "landscape"
+    if landscape:
+        w, h = h, w
+    layout = str(spec.get("layout", "hero")).lower()
+    if layout not in LAYOUTS:
+        warn(f"unknown layout {layout!r}; using hero")
+        layout = "hero"
+
+    headline = spec.get("headline") or die("spec needs a headline")
+    if len(headline.split()) > 10:
+        warn(f"headline is {len(headline.split())} words; ten or fewer read from across a room")
+    points = spec.get("points") or []
+    if len(points) > 4:
+        warn(f"{len(points)} points; a flyer carries four at most — the rest belong on the linked page")
+    body = spec.get("body") or ""
+    if len(body.split()) > 60:
+        warn(f"body is {len(body.split())} words; keep it under 60")
+    cta = spec.get("cta")
+    if isinstance(cta, list):
+        warn("more than one call to action; a flyer gets one — using the first")
+        cta = cta[0] if cta else None
+    if not cta:
+        warn("no call to action — what should a reader do next?")
+    img = spec.get("image") or {}
+    hero = data_uri(resolve(base, img.get("path")), "image") if isinstance(img, dict) else data_uri(resolve(base, img), "image")
+    if hero and isinstance(img, dict) and not img.get("alt"):
+        warn("image has no alt text")
+    qr = data_uri(resolve(base, spec.get("qr")), "QR code")
+    contact = {**(brand.get("contact") or {}), **(spec.get("contact") or {})}
+    footer = spec.get("footer") or (brand.get("legal") or {}).get("footer") or ""
+    details = spec.get("details") or []
+
+    def logo_img(dark):
+        src = logo_dark if dark else logo_light
+        return f'<img class="logo" src="{src}" alt="{esc(name)}">' if src else f'<span class="wordmark">{esc(name)}</span>'
+
+    pts = "".join(f"<li>{esc(p)}</li>" for p in points[:4])
+    det = "".join(f'<div class="det"><span>{esc(d.get("label"))}</span><b>{esc(d.get("value"))}</b></div>' for d in details)
+    cta_html = (f'<div class="cta"><b>{esc(cta.get("text"))}</b>'
+                + (f'<span>{esc(cta.get("url"))}</span>' if cta.get("url") else "")
+                + (f'<img class="qr" src="{qr}" alt="QR code for {esc(cta.get("url") or cta.get("text"))}">' if qr else "")
+                + "</div>") if cta else ""
+    contact_line = " · ".join(esc(v) for k, v in contact.items() if v and k in ("website", "email", "phone", "address"))
+    img_html = f'<div class="img" style="background-image:url({hero})" role="img" aria-label="{esc(img.get("alt", ""))}"></div>' if hero else ""
+
+    if layout == "split":
+        main = f"""<div class="split"><div class="side">{img_html or f'<div class="sideblock">{logo_img(True)}</div>'}</div>
+<div class="col">{'' if not img_html else logo_img(False)}<h1>{esc(headline)}</h1>
+{f'<p class="sub">{esc(spec.get("subhead"))}</p>' if spec.get("subhead") else ''}
+{f'<p class="body">{esc(body)}</p>' if body else ''}<ul class="pts">{pts}</ul><div class="dets">{det}</div>{cta_html}</div></div>"""
+    elif layout == "event":
+        first = details[0] if details else {}
+        main = f"""<header class="band">{logo_img(True)}</header>
+<div class="eventtop"><div class="date"><span>{esc(first.get("label", ""))}</span><b>{esc(first.get("value", ""))}</b></div>
+<div><h1>{esc(headline)}</h1>{f'<p class="sub">{esc(spec.get("subhead"))}</p>' if spec.get("subhead") else ''}</div></div>
+{img_html}<div class="col">{f'<p class="body">{esc(body)}</p>' if body else ''}<ul class="pts">{pts}</ul>
+<div class="dets">{"".join(f'<div class="det"><span>{esc(d.get("label"))}</span><b>{esc(d.get("value"))}</b></div>' for d in details[1:])}</div>{cta_html}</div>"""
+    elif layout == "minimal":
+        main = f"""<div class="col min">{logo_img(False)}<h1>{esc(headline)}</h1>
+{f'<p class="sub">{esc(spec.get("subhead"))}</p>' if spec.get("subhead") else ''}{img_html}
+{f'<p class="body">{esc(body)}</p>' if body else ''}<ul class="pts">{pts}</ul><div class="dets">{det}</div>{cta_html}</div>"""
+    else:  # hero
+        main = f"""<header class="band">{logo_img(True)}</header>{img_html}
+<div class="col"><h1>{esc(headline)}</h1>{f'<p class="sub">{esc(spec.get("subhead"))}</p>' if spec.get("subhead") else ''}
+{f'<p class="body">{esc(body)}</p>' if body else ''}<ul class="pts">{pts}</ul><div class="dets">{det}</div>{cta_html}</div>"""
+
+    unit = "px" if size in ("square", "story") else "pt"
+    # Social sizes are set in px. A square has a third less height than a page, so its type is
+    # smaller; a story is tall and takes larger type.
+    scale = {"square": 1.35, "story": 2.1}.get(size, 1.0)
+    doc = f"""<!doctype html>
+<html lang="en"><head><meta charset="utf-8">
+<title>{esc(headline)} — {esc(name)}</title>
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<style>
+@page {{ size: {w} {h}; margin: 0; }}
+* {{ box-sizing: border-box; }}
+html, body {{ margin: 0; padding: 0; background: #e9e9e6; }}
+body {{ -webkit-print-color-adjust: exact; print-color-adjust: exact; }}
+.page {{ width: {w}; height: {h}; margin: {'0' if unit == 'px' else '24px'} auto; background: {c['background']}; color: {c['text']};
+  font-family: {body_font}; position: relative; overflow: hidden; display: flex; flex-direction: column;
+  box-shadow: 0 6px 30px rgba(0,0,0,.18); font-size: {11 * scale}{unit}; line-height: 1.45; }}
+@media print {{ html, body {{ background: none; }} .page {{ margin: 0; box-shadow: none; }} }}
+h1 {{ font-family: {head_font}; color: {c['primary']}; font-size: {40 * scale}{unit}; line-height: 1.05; margin: 0 0 .35em; letter-spacing: -.01em; }}
+.sub {{ font-size: {16 * scale}{unit}; margin: 0 0 .9em; color: {c['secondary'] if contrast(c['secondary'], c['background']) >= 4.5 else c['text']}; }}
+.body {{ margin: 0 0 .9em; }}
+.band {{ background: {c['primary']}; color: {on_primary}; padding: {18 * scale}{unit} {36 * scale}{unit}; display: flex; align-items: center; }}
+.logo {{ max-height: {44 * scale}{unit}; max-width: 55%; }}
+.wordmark {{ font-family: {head_font}; font-weight: 700; font-size: {20 * scale}{unit}; }}
+.band .wordmark {{ color: {on_primary}; }}
+.img {{ height: {'26%' if size == 'square' else '34%'}; background-size: cover; background-position: center; }}
+.col {{ padding: {28 * scale}{unit} {36 * scale}{unit}; flex: 1; display: flex; flex-direction: column; }}
+.col .logo {{ margin-bottom: {18 * scale}{unit}; align-self: flex-start; }}
+.pts {{ list-style: none; padding: 0; margin: 0 0 1em; }}
+.pts li {{ padding-left: 1.2em; position: relative; margin: .35em 0; font-size: {13 * scale}{unit}; }}
+.pts li::before {{ content: ""; position: absolute; left: 0; top: .45em; width: .55em; height: .55em; background: {c['accent']}; border-radius: 2px; }}
+.dets {{ display: grid; grid-template-columns: repeat(auto-fit, minmax(30%, 1fr)); gap: .6em 1.2em; margin: 0 0 1.2em; }}
+.det span {{ display: block; font-size: {9 * scale}{unit}; text-transform: uppercase; letter-spacing: .08em; color: {c['secondary'] if contrast(c['secondary'], c['background']) >= 4.5 else c['text']}; }}
+.det b {{ font-size: {13 * scale}{unit}; }}
+.cta {{ margin-top: auto; background: {c['accent']}; color: {on_accent}; border-radius: 10px; padding: {16 * scale}{unit} {20 * scale}{unit};
+  display: grid; grid-template-columns: 1fr auto; align-items: center; gap: .2em 1em; }}
+.cta b {{ font-family: {head_font}; font-size: {20 * scale}{unit}; }}
+.cta span {{ grid-column: 1; font-size: {12 * scale}{unit}; }}
+.cta .qr {{ grid-column: 2; grid-row: 1 / span 2; width: {64 * scale}{unit}; height: {64 * scale}{unit}; background: #fff; padding: 4px; border-radius: 4px; }}
+.split {{ display: grid; grid-template-columns: 42% 1fr; flex: 1; }}
+.split .side .img {{ height: 100%; }}
+.sideblock {{ background: {c['primary']}; height: 100%; display: flex; align-items: flex-start; padding: {36 * scale}{unit} {28 * scale}{unit}; }}
+.eventtop {{ display: grid; grid-template-columns: auto 1fr; gap: 1.4em; align-items: center; padding: {28 * scale}{unit} {36 * scale}{unit} 0; }}
+.date {{ background: {c['accent']}; color: {on_accent}; border-radius: 10px; padding: .8em 1em; text-align: center; min-width: 7em; }}
+.date span {{ display: block; font-size: {9 * scale}{unit}; text-transform: uppercase; letter-spacing: .08em; }}
+.date b {{ font-family: {head_font}; font-size: {18 * scale}{unit}; line-height: 1.15; }}
+.eventtop + .img {{ margin-top: {20 * scale}{unit}; }}
+.min {{ justify-content: flex-start; }}
+.min .img {{ height: 30%; border-radius: 10px; margin: 0 0 1em; }}
+.foot {{ padding: {12 * scale}{unit} {36 * scale}{unit}; font-size: {9 * scale}{unit}; color: {c['text']}; border-top: 1px solid {c['secondary']}33; display: flex; justify-content: space-between; gap: 1em; }}
+</style></head>
+<body><main class="page">
+{main}
+<footer class="foot"><span>{contact_line}</span><span>{esc(footer)}</span></footer>
+</main></body></html>
+"""
+    out_path = Path(out_path).resolve()
+    out_path.parent.mkdir(parents=True, exist_ok=True)
+    out_path.write_text(doc, encoding="utf-8")
+    print(f"wrote {out_path} ({layout}, {size}{', landscape' if landscape else ''}; "
+          f"brand: {'yes' if brand else 'none'}; logo: {'yes' if (logo_light or logo_dark) else 'no'})")
+    print(f"  contrast: text {contrast(c['text'], c['background']):.1f}:1 · band {contrast(on_primary, c['primary']):.1f}:1 · "
+          f"call to action {contrast(on_accent, c['accent']):.1f}:1")
+    return out_path, size, landscape
+
+
+def chrome():
+    for p in ("/Applications/Google Chrome.app/Contents/MacOS/Google Chrome",
+              "/Applications/Chromium.app/Contents/MacOS/Chromium"):
+        if os.path.exists(p):
+            return p
+    for n in ("google-chrome", "google-chrome-stable", "chromium", "chromium-browser", "chrome"):
+        if shutil.which(n):
+            return shutil.which(n)
+    return None
+
+
+def export(html_path, size, pdf=None, png=None, landscape=False):
+    exe = chrome()
+    if not exe:
+        print("No Chrome or Chromium found: open the .html in a browser and print to PDF at 100% scale, margins none.")
+        return
+    url = Path(html_path).resolve().as_uri()
+    if pdf:
+        subprocess.run([exe, "--headless=new", "--disable-gpu", "--no-pdf-header-footer",
+                        f"--print-to-pdf={Path(pdf).resolve()}", url], check=False, capture_output=True)
+        print(f"wrote {Path(pdf).resolve()}" if Path(pdf).exists() else "PDF export failed — print from the browser instead")
+    if png:
+        pw, ph = PX[size][::-1] if landscape else PX[size]
+        subprocess.run([exe, "--headless=new", "--disable-gpu", "--hide-scrollbars", f"--window-size={pw},{ph}",
+                        f"--screenshot={Path(png).resolve()}", url], check=False, capture_output=True)
+        print(f"wrote {Path(png).resolve()}" if Path(png).exists() else "PNG export failed")
+
+
+def main(argv):
+    ap = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
+    ap.add_argument("spec")
+    ap.add_argument("-o", "--output", default=None, help="output .html (default: alongside the spec)")
+    ap.add_argument("--brand", default=None, help="brand.json from brand-kit; overrides the spec's brand field")
+    ap.add_argument("--pdf", default=None, help="also write a PDF through headless Chrome, when installed")
+    ap.add_argument("--png", default=None, help="also write a PNG (for social sizes) through headless Chrome")
+    a = ap.parse_args(argv)
+    out = a.output or str(Path(a.spec).with_suffix(".html"))
+    html_path, size, landscape = build(a.spec, out, a.brand)
+    if a.pdf or a.png:
+        export(html_path, size, a.pdf, a.png, landscape)
+    if WARNINGS:
+        print(f"{len(WARNINGS)} warning(s) — fix them or tell the user before sharing the flyer")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main(sys.argv[1:]))

package/packs/design/skills/marketing-flyer/templates/flyer-spec.json

@@ -0,0 +1,26 @@
+{
+  "$comment": "marketing-flyer spec. Paths are relative to this file. brand may be a path to brand.json (from brand-kit) or the same object inline; --brand on the command line overrides it. colors overrides individual brand colours for this flyer only.",
+  "brand": "./brand/brand.json",
+  "size": "letter",
+  "orientation": "portrait",
+  "layout": "event",
+  "headline": "Free community health fair",
+  "subhead": "Checkups, flu shots and healthy cooking — for the whole family",
+  "body": "",
+  "points": [
+    "Free blood pressure and glucose checks",
+    "Flu shots, no appointment needed",
+    "Kids' corner and healthy cooking demos"
+  ],
+  "details": [
+    { "label": "When", "value": "Sat 14 Nov, 10am–2pm" },
+    { "label": "Where", "value": "Riverside Park pavilion" },
+    { "label": "Cost", "value": "Free" }
+  ],
+  "cta": { "text": "Save your spot", "url": "northwindhealth.example/fair" },
+  "qr": "./qr-fair.png",
+  "image": { "path": "./riverside-park.jpg", "alt": "Families on the lawn at Riverside Park" },
+  "contact": { "phone": "" },
+  "footer": "© 2026 Northwind Health",
+  "colors": {}
+}

package/packs/dev-team/skills/contribution-wizard/SKILL.md

@@ -1,11 +1,11 @@
 ---
 name: contribution-wizard
-description: Guides an author through creating a new skilldrop skill from scratch — generates the manifest, SKILL.md, eval cases, and README entry from a plain-language description of the skill's purpose. Use when authoring a new skill for this repo or a private fork, ensuring a new skill meets the schema and validation requirements before submission.
+description: Guides an author through creating a new skilldrop skill from scratch — generates the manifest, SKILL.md, eval cases, and skill-catalogue entry from a plain-language description of the skill's purpose. Use when authoring a new skill for this repo or a private fork, ensuring a new skill meets the schema and validation requirements before submission.
 ---
 
 # contribution-wizard
 
-Walks an author through the full arc of creating a new skilldrop skill: intake → manifest + SKILL.md + evals + README entry. The output is copy-paste-ready — the author drops the files into `skills/<name>/` and runs `python3 validate.py` to confirm before opening a PR.
+Walks an author through the full arc of creating a new skilldrop skill: intake → manifest + SKILL.md + evals + skill-catalogue entry. The output is copy-paste-ready — the author drops the files into `packs/<pack>/skills/<name>/` and runs `python3 validate.py` to confirm before opening a PR.
 
 ## How to respond
 
@@ -89,12 +89,12 @@ Eight eval cases covering: 2 happy-path, 2 minimal/edge input, 2 anti-pattern (t
 ]
 ```
 
-#### `README.md` entry
+#### Skill catalogue entry
 
-One row for the skill table, matching the format of existing rows:
+One row for `guides/reference/skill-catalogue.md`, under the right category, matching the format of existing rows:
 
 ```
-| [`<skill-name>`](packs/<pack>/skills/<skill-name>/SKILL.md) | <One sentence describing what the skill does and the output it produces.> |
+| [`<skill-name>`](../../packs/<pack>/skills/<skill-name>/SKILL.md) | <One sentence describing what the skill does and the output it produces.> |
 ```
 
 #### Pack and outcome
@@ -110,7 +110,7 @@ Next steps:
 1. mkdir -p packs/<pack>/skills/<skill-name> && cd packs/<pack>/skills/<skill-name>
 2. Drop in manifest.json, SKILL.md, and evals/cases.json
 3. python3 validate.py    # must pass before opening a PR
-4. Add the README row to the correct section
+4. Add the row to guides/reference/skill-catalogue.md under the correct category
 5. Add the skill to an outcome in catalogue.json
 6. If the skill has a how-to guide, add it to guides/how-to/ and guides/README.md
 ```

package/packs/dev-team/skills/contribution-wizard/manifest.json

@@ -1,7 +1,7 @@
 {
   "name": "contribution-wizard",
-  "version": "1.0.1",
-  "description": "Guides an author through creating a new skilldrop skill from scratch — generates the manifest, SKILL.md, eval cases, and README entry from a plain-language description of the skill's purpose. Use when authoring a new skill for this repo or a private fork, ensuring a new skill meets the schema and validation requirements before submission.",
+  "version": "1.0.2",
+  "description": "Guides an author through creating a new skilldrop skill from scratch — generates the manifest, SKILL.md, eval cases, and skill-catalogue entry from a plain-language description of the skill's purpose. Use when authoring a new skill for this repo or a private fork, ensuring a new skill meets the schema and validation requirements before submission.",
   "entrypoint": "SKILL.md",
   "deps": { "npm": [], "pip": [] },
   "env": { "required": [], "optional": [] },

package/packs/stakeholder-comms/pack.json

@@ -1,5 +1,5 @@
 {
-  "description": "Non-technical audiences: audience profiling, deck outlines and real .pptx decks, exec summaries, decision logs, and guides.",
+  "description": "Non-technical audiences: audience profiling, exec summaries, decision logs, and guides. Decks live in the design pack.",
   "display_name": "Stakeholder comms",
   "keywords": ["exec-summary", "deck", "audience", "decision-log"],
   "links": {

package/profiles.json

@@ -16,7 +16,7 @@
     },
     "full": {
       "description": "The complete catalogue: every pack, every reviewer agent, every loop.",
-      "packs": ["core", "solution-architect", "product-manager", "dev-team", "sre-oncall", "stakeholder-comms", "ai-engineering", "claude-api"],
+      "packs": ["core", "solution-architect", "product-manager", "dev-team", "sre-oncall", "stakeholder-comms", "design", "ai-engineering", "claude-api"],
       "agents": ["devils-advocate", "security-reviewer", "code-quality"],
       "loops": ["discover", "design", "build", "release", "operate", "ship-a-draft"]
     }