feat(aaa): sync federation state — skills, specs, governance

- Skill patches: AGI-decisions-reflect, ASI-agent-invariants, FORGE-agentic-web-builder, FORGE-call-map, FORGE-route-least-power
- New specs: APEX_PATH_SELECTOR v0.2/v0.3, CONTRADICTION_FUNCTIONAL, ROUTING_FIELD
- Governance doctrine updates

DITEMPA BUKAN DIBERI

Author: kimi-code/FI-008

artifacts/aletta-ocean/aletta-ocean.html

@@ -0,0 +1,120 @@
+"""Render sado-reference page and capture full-page screenshot."""
+import asyncio
+from pathlib import Path
+from playwright.async_api import async_playwright
+
+URL = "https://syedos.arif-fazil.com/sado-reference/"
+OUT_DIR = Path("/root/AAA/artifacts/aletta-ocean")
+OUT_DIR.mkdir(parents=True, exist_ok=True)
+SCREENSHOT_PATH = OUT_DIR / "live-preview.png"
+
+async def main():
+    async with async_playwright() as p:
+        browser = await p.chromium.launch(headless=True)
+        context = await browser.new_context(
+            viewport={"width": 1440, "height": 900},
+            device_scale_factor=2,
+        )
+        page = await context.new_page()
+
+        console_msgs = []
+        page.on("console", lambda msg: console_msgs.append(f"[{msg.type}] {msg.text}"))
+        failed_requests = []
+        page.on("requestfailed", lambda req: failed_requests.append(f"{req.url} - {req.failure}"))
+
+        print(f"Navigating to {URL}...")
+        response = await page.goto(URL, wait_until="networkidle", timeout=30000)
+        print(f"Status: {response.status if response else 'no response'}")
+
+        # Wait 5s for fonts as specified
+        await page.wait_for_timeout(5000)
+
+        # Verify presence of expected elements
+        title_visible = await page.locator("text=ALETTA OCEAN").count()
+        vs_count = await page.locator("text=vs").count()
+        respect_section = await page.locator("text=Why Abang Sado Respect Dia").count()
+
+        # Image checks
+        imgs = page.locator("img")
+        img_count = await imgs.count()
+        img_details = []
+        for i in range(img_count):
+            img = imgs.nth(i)
+            src = await img.get_attribute("src")
+            alt = await img.get_attribute("alt")
+            natural_w = await img.evaluate("el => el.naturalWidth")
+            natural_h = await img.evaluate("el => el.naturalHeight")
+            complete = await img.evaluate("el => el.complete")
+            img_details.append({
+                "src": src, "alt": alt,
+                "naturalWidth": natural_w, "naturalHeight": natural_h,
+                "loaded": complete,
+            })
+
+        # Layout probe — get bounding boxes of portraits and vs circle
+        layout = await page.evaluate("""() => {
+            const result = {};
+            // Try to find portrait containers by alt or class
+            const allImgs = Array.from(document.querySelectorAll('img'));
+            result.imgCount = allImgs.length;
+            result.imgs = allImgs.map(i => ({
+                src: i.src,
+                alt: i.alt,
+                rect: i.getBoundingClientRect().toJSON(),
+            }));
+            // Look for 'vs' element
+            const allEls = Array.from(document.querySelectorAll('*'));
+            const vsEls = allEls.filter(e => e.textContent.trim() === 'vs' && e.children.length === 0);
+            result.vsElements = vsEls.map(v => ({
+                text: v.textContent,
+                tag: v.tagName,
+                rect: v.getBoundingClientRect().toJSON(),
+                classes: v.className,
+            }));
+            // Get page dimensions
+            result.pageHeight = document.documentElement.scrollHeight;
+            result.pageWidth = document.documentElement.scrollWidth;
+            result.viewportHeight = window.innerHeight;
+            result.viewportWidth = window.innerWidth;
+            result.title = document.title;
+            return result;
+        }""")
+
+        print("\n=== LAYOUT PROPOSURE ===")
+        print(f"Page title: {layout['title']}")
+        print(f"Page dimensions: {layout['pageWidth']}x{layout['pageHeight']} (viewport {layout['viewportWidth']}x{layout['viewportHeight']})")
+        print(f"Image count: {layout['imgCount']}")
+        for i, img in enumerate(layout['imgs']):
+            print(f"  img[{i}]: src={img['src'][-60:] if img['src'] else 'NONE'} alt={img['alt']} rect=({img['rect']['width']:.0f}x{img['rect']['height']:.0f} @ {img['rect']['x']:.0f},{img['rect']['y']:.0f})")
+        print(f"vs elements: {len(layout['vsElements'])}")
+        for v in layout['vsElements']:
+            print(f"  '{v['text']}' <{v['tag']}> class={v['classes']} rect=({v['rect']['width']:.0f}x{v['rect']['height']:.0f} @ {v['rect']['x']:.0f},{v['rect']['y']:.0f})")
+
+        print("\n=== TEXT PROPOSURE ===")
+        print(f"'ALETTA OCEAN' matches: {title_visible}")
+        print(f"'vs' matches: {vs_count}")
+        print(f"'Why Abang Sado Respect Dia' matches: {respect_section}")
+
+        print("\n=== IMAGES ===")
+        for img in img_details:
+            print(f"  src={img['src'][-60:] if img['src'] else 'NONE'} alt={img['alt']} {img['naturalWidth']}x{img['naturalHeight']} loaded={img['loaded']}")
+
+        if failed_requests:
+            print("\n=== FAILED REQUESTS ===")
+            for f in failed_requests:
+                print(f"  {f}")
+
+        if console_msgs:
+            print("\n=== CONSOLE (last 20) ===")
+            for m in console_msgs[-20:]:
+                print(f"  {m}")
+
+        # Capture full-page screenshot
+        print(f"\n=== CAPTURING FULL-PAGE SCREENSHOT to {SCREENSHOT_PATH} ===")
+        await page.screenshot(path=str(SCREENSHOT_PATH), full_page=True)
+        print(f"Screenshot saved: {SCREENSHOT_PATH}")
+        print(f"File size: {SCREENSHOT_PATH.stat().st_size} bytes")
+
+        await browser.close()
+
+asyncio.run(main())