feat: add --rich flag to convert clipboard Markdown to rich text (HTML)

README.md

@@ -110,7 +110,21 @@ clipdrop --ocr --auto-name       # Name a screenshot from its recognized text
 - The positional filename is optional; a provided extension is kept, otherwise format detection picks it
 - Falls back to a content-based slug when Apple Intelligence isn't available
 
-### 8. ✍️ **Transform** (macOS 26.0+) - Writing Tools for the Terminal
+### 8. 📋 **Rich Text Copy** - Paste Markdown into Confluence
+Copied Markdown pastes as raw `# syntax` in rich editors. `--rich` converts it
+to HTML and puts it back on the clipboard with both rich-text and plain-text
+flavors, so it pastes *formatted* into Confluence, Google Docs, Gmail, or Slack:
+```bash
+# Copy some markdown, then:
+clipdrop --rich                  # Convert → paste formatted anywhere
+clipdrop -r -p                   # Preview the HTML first
+clipdrop -r page                 # Also save it → page.html
+```
+- Headings, lists, tables, links, and code blocks survive the paste
+- Plain-text targets (and "Paste and Match Style") still get your original markdown
+- Code fence languages are mapped to what Confluence recognizes (`js` → `javascript`)
+
+### 9. ✍️ **Transform** (macOS 26.0+) - Writing Tools for the Terminal
 Reshape copied text on the way to a file, entirely on-device:
 ```bash
 clipdrop email.txt --rewrite formal                    # restyle (concise/friendly/…)
@@ -227,6 +241,7 @@ clipdrop -f <filename>   # Force overwrite
 ```bash
 clipdrop -yt            # YouTube transcript mode
 clipdrop --audio        # Force audio transcription
+clipdrop --rich         # Markdown → rich text, back to the clipboard
 ```
 
 ### Filters & Options

pyproject.toml

@@ -36,6 +36,7 @@ dependencies = [
     "requests>=2.34.2",
     "lxml>=6.1.1",
     "html2text>=2025.4.15",
+    "markdown-it-py>=3.0.0",
 ]
 
 [project.urls]

src/clipdrop/clipboard.py

@@ -1,5 +1,7 @@
 """Clipboard operations module for ClipDrop."""
 
+import subprocess
+import sys
 import time
 from typing import Optional, Dict, Any
 import pyperclip
@@ -27,6 +29,10 @@
 # Maximum content size (100MB)
 MAX_CONTENT_SIZE = 100 * 1024 * 1024
 
+# Maximum HTML size for rich text clipboard writes (10MB); hex encoding
+# doubles the osascript payload, so keep this well under MAX_CONTENT_SIZE
+MAX_RICH_CONTENT_SIZE = 10 * 1024 * 1024
+
 
 def get_text() -> Optional[str]:
     """
@@ -275,6 +281,95 @@ def copy_to_clipboard(content: str) -> None:
         raise ClipboardAccessError("Cannot copy to clipboard", original_error=e)
 
 
+def _escape_applescript_string(text: str) -> str:
+    """Escape text for embedding in a double-quoted AppleScript string."""
+    text = text.replace('\r\n', '\n')
+    text = text.replace('\\', '\\\\')
+    text = text.replace('"', '\\"')
+    text = text.replace('\n', '\\n')
+    text = text.replace('\r', '\\n')
+    text = text.replace('\t', '\\t')
+    return text
+
+
+def build_rich_clipboard_script(html: str, plain_text: str) -> str:
+    """
+    Build an AppleScript that sets both HTML and plain-text clipboard flavors.
+
+    The HTML is hex-encoded into a «data HTML…» literal (the inverse of the
+    read path in html_parser.get_html_from_clipboard), so it needs no
+    escaping. The text attribute supplies the plain-text flavor — without it
+    paste breaks entirely in apps like Slack and Firefox.
+
+    Args:
+        html: HTML fragment for the rich text flavor
+        plain_text: Plain-text fallback (typically the original markdown)
+
+    Returns:
+        AppleScript source string
+    """
+    hex_html = html.encode('utf-8').hex().upper()
+    escaped_text = _escape_applescript_string(plain_text)
+    return (
+        f'set the clipboard to '
+        f'{{text:"{escaped_text}", «class HTML»:«data HTML{hex_html}»}}'
+    )
+
+
+def copy_rich_text_to_clipboard(html: str, plain_text: str) -> None:
+    """
+    Copy rich text (HTML) to the macOS clipboard with a plain-text fallback.
+
+    Sets both the public.html and plain-text pasteboard flavors so rich
+    editors (Confluence, Google Docs, Gmail, Slack) paste formatted content
+    while plain-text targets get the original source.
+
+    Args:
+        html: HTML fragment for the rich text flavor
+        plain_text: Plain-text fallback (typically the original markdown)
+
+    Raises:
+        ContentTooLargeError: If HTML exceeds MAX_RICH_CONTENT_SIZE
+        ClipboardAccessError: If not on macOS or osascript fails
+    """
+    html_size = len(html.encode('utf-8'))
+    if html_size > MAX_RICH_CONTENT_SIZE:
+        raise ContentTooLargeError(html_size, MAX_RICH_CONTENT_SIZE)
+
+    if sys.platform != 'darwin':
+        raise ClipboardAccessError(
+            "Rich text clipboard copy requires macOS"
+        )
+
+    script = build_rich_clipboard_script(html, plain_text)
+    try:
+        # Script over stdin: hex-encoded HTML can exceed argv size limits
+        result = subprocess.run(
+            ['osascript', '-'],
+            input=script.encode('utf-8'),
+            capture_output=True,
+            timeout=10
+        )
+    except subprocess.TimeoutExpired as e:
+        raise ClipboardAccessError(
+            "Timed out writing rich text to clipboard", original_error=e
+        )
+    except OSError as e:
+        raise ClipboardAccessError(
+            "Cannot run osascript to write rich text", original_error=e
+        )
+
+    if result.returncode != 0:
+        stderr = result.stderr.decode('utf-8', errors='ignore').strip()
+        raise ClipboardAccessError(
+            f"Failed to write rich text to clipboard: {stderr or 'osascript error'}"
+        )
+
+    # Update cache with the plain-text flavor
+    _clipboard_cache['content'] = plain_text
+    _clipboard_cache['timestamp'] = time.time()
+
+
 # Image clipboard operations
 _image_cache: Dict[str, Any] = {
     'image': None,

src/clipdrop/main.py

@@ -17,7 +17,7 @@
 from rich.prompt import Confirm
 
 from clipdrop import __version__
-from clipdrop import clipboard, detect, files, images, pdf
+from clipdrop import clipboard, detect, files, images, pdf, richtext
 from clipdrop.macos_ai import summarize_content, summarize_content_with_chunking
 from clipdrop.error_helpers import display_error, show_success_message
 from clipdrop.paranoid import (
@@ -26,6 +26,8 @@
     print_binary_skip_notice,
 )
 from clipdrop.exceptions import (
+    ClipboardAccessError,
+    ContentTooLargeError,
     YTDLPNotFoundError,
     NoCaptionsError,
     YouTubeError
@@ -1039,6 +1041,87 @@ def handle_youtube_transcript(
         raise typer.Exit(1)
 
 
+def handle_rich_copy(
+    filename: Optional[str],
+    scan: bool,
+    paranoid_mode: Optional[ParanoidMode],
+    preview: bool,
+    force: bool,
+    yes: bool,
+) -> None:
+    """Convert clipboard Markdown to rich text (HTML) and copy it back.
+
+    Sets both HTML and plain-text clipboard flavors so pasting into rich
+    editors (Confluence, Google Docs, Gmail, Slack) produces formatted
+    content while plain-text targets still get the original markdown.
+    Optionally saves the HTML to a file when a filename is given.
+    """
+    content = clipboard.get_text()
+    if not content or not content.strip():
+        display_error('empty_clipboard')
+        raise typer.Exit(1)
+
+    if not detect.is_markdown(content):
+        console.print(
+            "[yellow]⚠️ Clipboard content doesn't look like Markdown — "
+            "converting anyway[/yellow]"
+        )
+
+    # Scan the markdown source before it goes anywhere
+    active_paranoid = paranoid_mode or (ParanoidMode.PROMPT if scan else None)
+    if active_paranoid:
+        content, _ = paranoid_gate(
+            content,
+            active_paranoid,
+            is_tty=sys.stdin.isatty(),
+            auto_yes=yes
+        )
+        if content is None:
+            console.print("[yellow]⚠️ Content not copied (paranoid mode)[/yellow]")
+            raise typer.Exit(0)
+
+    try:
+        html = richtext.markdown_to_rich_html(content)
+    except ValueError as e:
+        console.print(f"[red]❌ {str(e)}[/red]")
+        raise typer.Exit(1)
+
+    if preview:
+        console.print(Panel(
+            Syntax(html, "html", word_wrap=True),
+            title="Rich text (HTML) preview",
+            border_style="cyan"
+        ))
+        if not yes and not Confirm.ask(
+            "\n[yellow]Copy rich text to clipboard?[/yellow]"
+        ):
+            console.print("[yellow]Cancelled[/yellow]")
+            raise typer.Exit(0)
+
+    try:
+        clipboard.copy_rich_text_to_clipboard(html, plain_text=content)
+    except (ContentTooLargeError, ClipboardAccessError) as e:
+        console.print(f"[red]❌ {str(e)}[/red]")
+        raise typer.Exit(1)
+
+    if filename:
+        output_filename = filename if Path(filename).suffix else f"{filename}.html"
+        files.write_text(output_filename, html, force=force)
+        console.print(f"[green]💾 HTML saved to '{output_filename}'[/green]")
+
+    html_size = len(html.encode('utf-8'))
+    size_str = (
+        f"{html_size:,} bytes" if html_size < 1024 else f"{html_size/1024:.1f} KB"
+    )
+    console.print(Panel(
+        "[green]✅ Rich text copied to clipboard[/green]\n"
+        "Paste into Confluence, Google Docs, Gmail, or Slack for "
+        "formatted content.\n"
+        f"[dim]HTML: {size_str} | Plain-text fallback: original markdown[/dim]",
+        border_style="green"
+    ))
+
+
 def main(
     filename: Optional[str] = typer.Argument(
         None,
@@ -1170,6 +1253,15 @@ def main(
              "Outputs: .srt (subtitles), .txt (plain text), or .md (markdown). "
              "Requires macOS 26.0+ with Apple Intelligence"
     ),
+    rich: bool = typer.Option(
+        False,
+        "--rich",
+        "-r",
+        help="Convert clipboard Markdown to rich text (HTML) and copy it "
+             "back to the clipboard for pasting into Confluence, Google "
+             "Docs, Gmail, or Slack. Saves the HTML too when a filename "
+             "is given"
+    ),
     version: Optional[bool] = typer.Option(
         None,
         "--version",
@@ -1208,6 +1300,11 @@ def main(
         clipdrop -yt --lang es      # Spanish transcript
         clipdrop -yt --chapters     # Include chapter markers
 
+      [green]Rich Text:[/green]
+        clipdrop --rich             # Markdown → rich text, back to clipboard
+        clipdrop -r -p              # Preview HTML before copying
+        clipdrop -r page            # Also save the HTML → page.html
+
       [green]Mixed Content:[/green]
         clipdrop document           # Mixed text+image → document.pdf
         clipdrop content --text-only # Forces text mode
@@ -1263,6 +1360,18 @@ def main(
             summarize=summarize,
         )
 
+    # Check if this is rich text mode (before audio auto-detect so stale
+    # clipboard audio can't hijack the command)
+    if rich:
+        return handle_rich_copy(
+            filename=filename,
+            scan=paranoid_flag,
+            paranoid_mode=paranoid_mode,
+            preview=preview,
+            force=force,
+            yes=yes,
+        )
+
     # Check for audio in clipboard (with or without filename)
     try:
         from clipdrop.macos_ai import check_audio_in_clipboard
@@ -1288,6 +1397,7 @@ def main(
         console.print("  clipdrop data.json     # Save JSON")
         console.print("  clipdrop --youtube     # Download YouTube transcript")
         console.print("  clipdrop --audio       # Transcribe audio from clipboard")
+        console.print("  clipdrop --rich        # Markdown → rich text clipboard")
         console.print("  clipdrop -yt output.srt # YouTube with custom name")
         console.print("\n[dim]Try 'clipdrop --help' for more options[/dim]")
         raise typer.Exit(1)