diff --git a/skills/publish-to-pages/SKILL.md b/skills/publish-to-pages/SKILL.md
new file mode 100644
index 000000000..a13a12864
--- /dev/null
+++ b/skills/publish-to-pages/SKILL.md
@@ -0,0 +1,90 @@
+---
+name: publish-to-pages
+description: 'Publish presentations and web content to GitHub Pages. Converts PPTX, PDF, HTML, or Google Slides to a live GitHub Pages URL. Handles repo creation, file conversion, Pages enablement, and returns the live URL. Use when the user wants to publish, deploy, or share a presentation or HTML file via GitHub Pages.'
+---
+
+# publish-to-pages
+
+Publish any presentation or web content to GitHub Pages in one shot.
+
+## 1. Prerequisites Check
+
+Run these silently. Only surface errors:
+
+```bash
+command -v gh >/dev/null || echo "MISSING: gh CLI — install from https://cli.github.com"
+gh auth status &>/dev/null || echo "MISSING: gh not authenticated — run 'gh auth login'"
+command -v python3 >/dev/null || echo "MISSING: python3 (needed for PPTX conversion)"
+```
+
+`poppler-utils` is optional (PDF conversion via `pdftoppm`). Don't block on it.
+
+## 2. Input Detection
+
+Determine input type from what the user provides:
+
+| Input | Detection |
+|-------|-----------|
+| HTML file | Extension `.html` or `.htm` |
+| PPTX file | Extension `.pptx` |
+| PDF file | Extension `.pdf` |
+| Google Slides URL | URL contains `docs.google.com/presentation` |
+
+Ask the user for a **repo name** if not provided. Default: filename without extension.
+
+## 3. Conversion
+
+### HTML
+No conversion needed. Use the file directly as `index.html`.
+
+### PPTX
+Run the conversion script:
+```bash
+python3 SKILL_DIR/scripts/convert-pptx.py INPUT_FILE /tmp/output.html
+```
+If `python-pptx` is missing, tell the user: `pip install python-pptx`
+
+### PDF
+Convert with the included script (requires `poppler-utils` for `pdftoppm`):
+```bash
+python3 SKILL_DIR/scripts/convert-pdf.py INPUT_FILE /tmp/output.html
+```
+Each page is rendered as a PNG and base64-embedded into a self-contained HTML with slide navigation.
+If `pdftoppm` is missing, tell the user: `apt install poppler-utils` (or `brew install poppler` on macOS).
+
+### Google Slides
+1. Extract the presentation ID from the URL (the long string between `/d/` and `/`)
+2. Download as PPTX:
+```bash
+curl -L "https://docs.google.com/presentation/d/PRESENTATION_ID/export/pptx" -o /tmp/slides.pptx
+```
+3. Then convert the PPTX using the convert script above.
+
+## 4. Publishing
+
+### Visibility
+Repos are created **public** by default. If the user specifies `private` (or wants a private repo), use `--private` — but note that GitHub Pages on private repos requires a Pro, Team, or Enterprise plan.
+
+### Publish
+```bash
+bash SKILL_DIR/scripts/publish.sh /path/to/index.html REPO_NAME public "Description"
+```
+
+Pass `private` instead of `public` if the user requests it.
+
+The script creates the repo, pushes `index.html`, and enables GitHub Pages.
+
+## 5. Output
+
+Tell the user:
+- **Repository:** `https://github.com/USERNAME/REPO_NAME`
+- **Live URL:** `https://USERNAME.github.io/REPO_NAME/`
+- **Note:** Pages takes 1-2 minutes to go live.
+
+## Error Handling
+
+- **Repo already exists:** Suggest appending a number (`my-slides-2`) or a date (`my-slides-2026`).
+- **Pages enablement fails:** Still return the repo URL. User can enable Pages manually in repo Settings.
+- **PPTX conversion fails:** Tell user to run `pip install python-pptx`.
+- **PDF conversion fails:** Suggest installing `poppler-utils` (`apt install poppler-utils` or `brew install poppler`).
+- **Google Slides download fails:** The presentation may not be publicly accessible. Ask user to make it viewable or download the PPTX manually.
diff --git a/skills/publish-to-pages/scripts/convert-pdf.py b/skills/publish-to-pages/scripts/convert-pdf.py
new file mode 100755
index 000000000..01c99e739
--- /dev/null
+++ b/skills/publish-to-pages/scripts/convert-pdf.py
@@ -0,0 +1,121 @@
+#!/usr/bin/env python3
+"""Convert a PDF to a self-contained HTML presentation.
+
+Each page is rendered as a PNG image (via pdftoppm) and base64-embedded
+into a single HTML file with slide navigation (arrows, swipe, click).
+
+Requirements: poppler-utils (pdftoppm)
+Usage: python3 convert-pdf.py input.pdf [output.html]
+"""
+
+import base64
+import glob
+import os
+import subprocess
+import sys
+import tempfile
+from pathlib import Path
+
+
+def convert(pdf_path: str, output_path: str | None = None, dpi: int = 150):
+    pdf_path = str(Path(pdf_path).resolve())
+    if not Path(pdf_path).exists():
+        print(f"Error: {pdf_path} not found")
+        sys.exit(1)
+
+    # Check for pdftoppm
+    if subprocess.run(["which", "pdftoppm"], capture_output=True).returncode != 0:
+        print("Error: pdftoppm not found. Install poppler-utils:")
+        print("  apt install poppler-utils  # Debian/Ubuntu")
+        print("  brew install poppler       # macOS")
+        sys.exit(1)
+
+    with tempfile.TemporaryDirectory() as tmpdir:
+        prefix = os.path.join(tmpdir, "page")
+        result = subprocess.run(
+            ["pdftoppm", "-png", "-r", str(dpi), pdf_path, prefix],
+            capture_output=True, text=True
+        )
+        if result.returncode != 0:
+            print(f"Error converting PDF: {result.stderr}")
+            sys.exit(1)
+
+        pages = sorted(glob.glob(f"{prefix}-*.png"))
+        if not pages:
+            print("Error: No pages rendered from PDF")
+            sys.exit(1)
+
+        slides_html = []
+        for i, page_path in enumerate(pages, 1):
+            with open(page_path, "rb") as f:
+                b64 = base64.b64encode(f.read()).decode()
+            slides_html.append(
+                f'<section class="slide">'
+                f'<div class="slide-inner">'
+                f'<img src="data:image/png;base64,{b64}" alt="Page {i}">'
+                f'</div></section>'
+            )
+
+    # Try to extract title from filename
+    title = Path(pdf_path).stem.replace("-", " ").replace("_", " ")
+
+    html = f'''<!DOCTYPE html>
+<html lang="en">
+<head>
+<meta charset="UTF-8">
+<meta name="viewport" content="width=device-width, initial-scale=1.0">
+<title>{title}</title>
+<style>
+* {{ margin: 0; padding: 0; box-sizing: border-box; }}
+html, body {{ height: 100%; overflow: hidden; background: #000; }}
+.slide {{ width: 100vw; height: 100vh; display: none; align-items: center; justify-content: center; }}
+.slide.active {{ display: flex; }}
+.slide-inner {{ display: flex; align-items: center; justify-content: center; width: 100%; height: 100%; }}
+.slide-inner img {{ max-width: 100%; max-height: 100%; object-fit: contain; }}
+.progress {{ position: fixed; bottom: 0; left: 0; height: 4px; background: #0366d6; transition: width 0.3s; z-index: 100; }}
+.counter {{ position: fixed; bottom: 12px; right: 20px; font-size: 14px; color: rgba(255,255,255,0.4); z-index: 100; }}
+</style>
+</head>
+<body>
+{chr(10).join(slides_html)}
+<div class="progress" id="progress"></div>
+<div class="counter" id="counter"></div>
+<script>
+const slides = document.querySelectorAll('.slide');
+let current = 0;
+function show(n) {{
+    slides.forEach(s => s.classList.remove('active'));
+    current = Math.max(0, Math.min(n, slides.length - 1));
+    slides[current].classList.add('active');
+    document.getElementById('progress').style.width = ((current + 1) / slides.length * 100) + '%';
+    document.getElementById('counter').textContent = (current + 1) + ' / ' + slides.length;
+}}
+document.addEventListener('keydown', e => {{
+    if (e.key === 'ArrowRight' || e.key === ' ') {{ e.preventDefault(); show(current + 1); }}
+    if (e.key === 'ArrowLeft') {{ e.preventDefault(); show(current - 1); }}
+}});
+let touchStartX = 0;
+document.addEventListener('touchstart', e => {{ touchStartX = e.changedTouches[0].screenX; }});
+document.addEventListener('touchend', e => {{
+    const diff = e.changedTouches[0].screenX - touchStartX;
+    if (Math.abs(diff) > 50) {{ diff > 0 ? show(current - 1) : show(current + 1); }}
+}});
+document.addEventListener('click', e => {{
+    if (e.clientX > window.innerWidth / 2) show(current + 1);
+    else show(current - 1);
+}});
+show(0);
+</script>
+</body></html>'''
+
+    output = output_path or str(Path(pdf_path).with_suffix('.html'))
+    Path(output).write_text(html, encoding='utf-8')
+    print(f"Converted to: {output}")
+    print(f"Pages: {len(slides_html)}")
+
+
+if __name__ == "__main__":
+    if len(sys.argv) < 2:
+        print("Usage: convert-pdf.py <file.pdf> [output.html]")
+        sys.exit(1)
+    convert(sys.argv[1], sys.argv[2] if len(sys.argv) > 2 else None)
diff --git a/skills/publish-to-pages/scripts/convert-pptx.py b/skills/publish-to-pages/scripts/convert-pptx.py
new file mode 100755
index 000000000..6423b14af
--- /dev/null
+++ b/skills/publish-to-pages/scripts/convert-pptx.py
@@ -0,0 +1,306 @@
+#!/usr/bin/env python3
+"""Convert a PPTX file to a self-contained HTML presentation with formatting preserved."""
+import sys
+import base64
+import io
+import re
+from pathlib import Path
+
+try:
+    from pptx import Presentation
+    from pptx.util import Inches, Pt, Emu
+    from pptx.enum.text import PP_ALIGN
+    from pptx.dml.color import RGBColor
+except ImportError:
+    print("ERROR: python-pptx not installed. Install with: pip install python-pptx")
+    sys.exit(1)
+
+
+def rgb_to_hex(rgb_color):
+    """Convert RGBColor to hex string."""
+    if rgb_color is None:
+        return None
+    try:
+        return f"#{rgb_color}"
+    except:
+        return None
+
+
+def get_text_style(run):
+    """Extract inline text styling from a run."""
+    styles = []
+    try:
+        if run.font.bold:
+            styles.append("font-weight:bold")
+        if run.font.italic:
+            styles.append("font-style:italic")
+        if run.font.underline:
+            styles.append("text-decoration:underline")
+        if run.font.size:
+            styles.append(f"font-size:{run.font.size.pt}pt")
+        if run.font.color and run.font.color.rgb:
+            styles.append(f"color:{rgb_to_hex(run.font.color.rgb)}")
+        if run.font.name:
+            styles.append(f"font-family:'{run.font.name}',sans-serif")
+    except:
+        pass
+    return ";".join(styles)
+
+
+def get_alignment(paragraph):
+    """Get CSS text-align from paragraph alignment."""
+    try:
+        align = paragraph.alignment
+        if align == PP_ALIGN.CENTER:
+            return "center"
+        elif align == PP_ALIGN.RIGHT:
+            return "right"
+        elif align == PP_ALIGN.JUSTIFY:
+            return "justify"
+    except:
+        pass
+    return "left"
+
+
+def extract_image(shape):
+    """Extract image from shape as base64 data URI."""
+    try:
+        image = shape.image
+        content_type = image.content_type
+        image_bytes = image.blob
+        b64 = base64.b64encode(image_bytes).decode('utf-8')
+        return f"data:{content_type};base64,{b64}"
+    except:
+        return None
+
+
+def get_shape_position(shape, slide_width, slide_height):
+    """Get shape position as percentages."""
+    try:
+        left = (shape.left / slide_width) * 100 if shape.left else 0
+        top = (shape.top / slide_height) * 100 if shape.top else 0
+        width = (shape.width / slide_width) * 100 if shape.width else 50
+        height = (shape.height / slide_height) * 100 if shape.height else 30
+        return left, top, width, height
+    except:
+        return 5, 5, 90, 40
+
+
+def get_slide_background(slide, prs):
+    """Extract slide background color from XML."""
+    from pptx.oxml.ns import qn
+    for source in [slide, slide.slide_layout]:
+        try:
+            bg_el = source.background._element
+            # Look for solidFill > srgbClr inside bgPr
+            for sf in bg_el.iter(qn('a:solidFill')):
+                clr = sf.find(qn('a:srgbClr'))
+                if clr is not None and clr.get('val'):
+                    return f"background-color:#{clr.get('val')}"
+        except:
+            pass
+    return "background-color:#ffffff"
+
+
+def get_shape_fill(shape):
+    """Extract shape fill color from XML."""
+    from pptx.oxml.ns import qn
+    try:
+        sp_pr = shape._element.find(qn('p:spPr'))
+        if sp_pr is None:
+            sp_pr = shape._element.find(qn('a:spPr'))
+        if sp_pr is None:
+            # Try direct child
+            for tag in ['{http://schemas.openxmlformats.org/drawingml/2006/main}spPr',
+                        '{http://schemas.openxmlformats.org/presentationml/2006/main}spPr']:
+                sp_pr = shape._element.find(tag)
+                if sp_pr is not None:
+                    break
+        if sp_pr is not None:
+            sf = sp_pr.find(qn('a:solidFill'))
+            if sf is not None:
+                clr = sf.find(qn('a:srgbClr'))
+                if clr is not None and clr.get('val'):
+                    return f"#{clr.get('val')}"
+    except:
+        pass
+    return None
+
+
+def render_paragraph(paragraph):
+    """Render a paragraph with inline formatting."""
+    align = get_alignment(paragraph)
+    parts = []
+    for run in paragraph.runs:
+        text = run.text
+        if not text:
+            continue
+        text = text.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;")
+        style = get_text_style(run)
+        if style:
+            parts.append(f'<span style="{style}">{text}</span>')
+        else:
+            parts.append(text)
+    if not parts:
+        return ""
+    content = "".join(parts)
+    return f'<p style="text-align:{align};margin:0.3em 0;line-height:1.4">{content}</p>'
+
+
+def convert(pptx_path, output_path=None):
+    prs = Presentation(pptx_path)
+    slide_width = prs.slide_width
+    slide_height = prs.slide_height
+    aspect_ratio = slide_width / slide_height if slide_height else 16/9
+
+    slides_html = []
+
+    for i, slide in enumerate(prs.slides, 1):
+        bg_style = get_slide_background(slide, prs)
+        elements = []
+
+        for shape in sorted(slide.shapes, key=lambda s: (s.top or 0, s.left or 0)):
+            left, top, width, height = get_shape_position(shape, slide_width, slide_height)
+            pos_style = f"position:absolute;left:{left:.1f}%;top:{top:.1f}%;width:{width:.1f}%;height:{height:.1f}%"
+
+            # Image
+            if shape.shape_type == 13 or hasattr(shape, "image"):
+                data_uri = extract_image(shape)
+                if data_uri:
+                    elements.append(
+                        f'<div style="{pos_style};display:flex;align-items:center;justify-content:center">'
+                        f'<img src="{data_uri}" style="max-width:100%;max-height:100%;object-fit:contain" alt="">'
+                        f'</div>'
+                    )
+                    continue
+
+            # Table
+            if shape.has_table:
+                table = shape.table
+                table_html = '<table style="width:100%;border-collapse:collapse;font-size:0.9em">'
+                for row in table.rows:
+                    table_html += "<tr>"
+                    for cell in row.cells:
+                        cell_text = cell.text.replace("&", "&amp;").replace("<", "&lt;")
+                        table_html += f'<td style="border:1px solid #ccc;padding:6px 10px">{cell_text}</td>'
+                    table_html += "</tr>"
+                table_html += "</table>"
+                elements.append(f'<div style="{pos_style};overflow:auto">{table_html}</div>')
+                continue
+
+            # Text
+            if shape.has_text_frame:
+                text_parts = []
+                for para in shape.text_frame.paragraphs:
+                    rendered = render_paragraph(para)
+                    if rendered:
+                        text_parts.append(rendered)
+                if text_parts:
+                    content = "".join(text_parts)
+                    fill = get_shape_fill(shape)
+                    fill_style = f"background-color:{fill};padding:1em;border-radius:8px;" if fill else ""
+                    elements.append(
+                        f'<div style="{pos_style};{fill_style}overflow:hidden;display:flex;flex-direction:column;justify-content:center">'
+                        f'{content}</div>'
+                    )
+                continue
+
+            # Decorative shape with fill (colored rectangles, bars, etc.)
+            fill = get_shape_fill(shape)
+            if fill:
+                elements.append(
+                    f'<div style="{pos_style};background-color:{fill};border-radius:4px"></div>'
+                )
+
+        slide_content = "\n".join(elements)
+        slides_html.append(
+            f'<section class="slide" style="{bg_style}">\n<div class="slide-inner">\n{slide_content}\n</div>\n</section>'
+        )
+
+    title = "Presentation"
+    # Try to get title from first slide
+    if prs.slides:
+        for shape in prs.slides[0].shapes:
+            if hasattr(shape, "text") and shape.text.strip() and len(shape.text.strip()) < 150:
+                title = shape.text.strip()
+                break
+
+    html = f'''<!DOCTYPE html>
+<html lang="en">
+<head>
+<meta charset="UTF-8">
+<meta name="viewport" content="width=device-width, initial-scale=1.0">
+<title>{title}</title>
+<style>
+* {{ margin: 0; padding: 0; box-sizing: border-box; }}
+html, body {{ height: 100%; overflow: hidden; background: #000; }}
+.slide {{
+    width: 100vw; height: 100vh;
+    display: none;
+    align-items: center; justify-content: center;
+    overflow: hidden;
+}}
+.slide.active {{ display: flex; }}
+.slide-inner {{
+    position: relative;
+    width: 1280px; height: {int(1280 / aspect_ratio)}px;
+    transform-origin: center center;
+    flex-shrink: 0;
+}}
+.progress {{ position: fixed; bottom: 0; left: 0; height: 4px; background: #0366d6; transition: width 0.3s; z-index: 100; }}
+.counter {{ position: fixed; bottom: 12px; right: 20px; font-size: 14px; color: rgba(255,255,255,0.4); z-index: 100; }}
+</style>
+</head>
+<body>
+{chr(10).join(slides_html)}
+<div class="progress" id="progress"></div>
+<div class="counter" id="counter"></div>
+<script>
+const slides = document.querySelectorAll('.slide');
+let current = 0;
+function show(n) {{
+    slides.forEach(s => s.classList.remove('active'));
+    current = Math.max(0, Math.min(n, slides.length - 1));
+    slides[current].classList.add('active');
+    document.getElementById('progress').style.width = ((current + 1) / slides.length * 100) + '%';
+    document.getElementById('counter').textContent = (current + 1) + ' / ' + slides.length;
+}}
+document.addEventListener('keydown', e => {{
+    if (e.key === 'ArrowRight' || e.key === ' ') {{ e.preventDefault(); show(current + 1); }}
+    if (e.key === 'ArrowLeft') {{ e.preventDefault(); show(current - 1); }}
+}});
+let touchStartX = 0;
+document.addEventListener('touchstart', e => {{ touchStartX = e.changedTouches[0].screenX; }});
+document.addEventListener('touchend', e => {{
+    const diff = e.changedTouches[0].screenX - touchStartX;
+    if (Math.abs(diff) > 50) {{ diff > 0 ? show(current - 1) : show(current + 1); }}
+}});
+document.addEventListener('click', e => {{
+    if (e.clientX > window.innerWidth / 2) show(current + 1);
+    else show(current - 1);
+}});
+show(0);
+function scaleSlides() {{
+    document.querySelectorAll('.slide-inner').forEach(inner => {{
+        const scaleX = window.innerWidth / inner.offsetWidth;
+        const scaleY = window.innerHeight / inner.offsetHeight;
+        const scale = Math.min(scaleX, scaleY);
+        inner.style.transform = 'scale(' + scale + ')';
+    }});
+}}
+window.addEventListener('resize', scaleSlides);
+scaleSlides();
+</script>
+</body></html>'''
+
+    output = output_path or str(Path(pptx_path).with_suffix('.html'))
+    Path(output).write_text(html, encoding='utf-8')
+    print(f"Converted to: {output}")
+    print(f"Slides: {len(slides_html)}")
+
+
+if __name__ == "__main__":
+    if len(sys.argv) < 2:
+        print("Usage: convert-pptx.py <file.pptx> [output.html]")
+        sys.exit(1)
+    convert(sys.argv[1], sys.argv[2] if len(sys.argv) > 2 else None)
diff --git a/skills/publish-to-pages/scripts/publish.sh b/skills/publish-to-pages/scripts/publish.sh
new file mode 100755
index 000000000..ddebfc6fc
--- /dev/null
+++ b/skills/publish-to-pages/scripts/publish.sh
@@ -0,0 +1,40 @@
+#!/bin/bash
+# Main publish script
+# Args: $1 = path to index.html, $2 = repo name, $3 = visibility (private|public), $4 = description
+set -euo pipefail
+
+HTML_FILE="$1"
+REPO_NAME="$2"
+VISIBILITY="${3:-public}"
+DESCRIPTION="${4:-Published via publish-to-pages}"
+
+USERNAME=$(gh api user --jq '.login')
+
+# Check if repo exists
+if gh repo view "$USERNAME/$REPO_NAME" &>/dev/null; then
+    echo "ERROR: Repository $USERNAME/$REPO_NAME already exists"
+    exit 1
+fi
+
+# Create repo
+gh repo create "$REPO_NAME" --"$VISIBILITY" --description "$DESCRIPTION"
+
+# Clone, push, enable pages
+TMPDIR=$(mktemp -d)
+git clone "https://github.com/$USERNAME/$REPO_NAME.git" "$TMPDIR"
+cp "$HTML_FILE" "$TMPDIR/index.html"
+cd "$TMPDIR"
+git add index.html
+git commit -m "Publish content"
+git push origin main
+
+# Enable GitHub Pages
+gh api "repos/$USERNAME/$REPO_NAME/pages" -X POST -f source[branch]=main -f source[path]=/ 2>/dev/null || true
+
+echo "REPO_URL=https://github.com/$USERNAME/$REPO_NAME"
+echo "PAGES_URL=https://$USERNAME.github.io/$REPO_NAME/"
+echo ""
+echo "GitHub Pages may take 1-2 minutes to deploy."
+
+# Cleanup
+rm -rf "$TMPDIR"