devendra-skill-package 1.0.2

package/.github/skills/md-file-converter/extractors/extract-pdf.sh

@@ -0,0 +1,97 @@
+#!/usr/bin/env bash
+# PDF extractor. Tries pdftotext → pandoc → PyPDF2 in order.
+# Usage: bash extractors/extract-pdf.sh <file.pdf> [start_page] [end_page]
+# start_page and end_page are optional (1-based). Omit for full document.
+
+FILE="$1"
+START="${2:-}"
+END="${3:-}"
+
+if [ -z "$FILE" ]; then
+  echo "Usage: extract-pdf.sh <file.pdf> [start_page] [end_page]"
+  exit 1
+fi
+
+if [ ! -f "$FILE" ]; then
+  echo "Error: File not found: $FILE"
+  exit 1
+fi
+
+echo "=== PDF Extraction ==="
+echo "File     : $FILE"
+[ -n "$START" ] && echo "Pages    : $START to ${END:-end}"
+echo ""
+
+# Strategy 1: pdftotext (poppler-utils) — preserves layout best
+if command -v pdftotext >/dev/null 2>&1; then
+  echo "Method: pdftotext"
+  echo ""
+  echo "--- Content ---"
+  if [ -n "$START" ] && [ -n "$END" ]; then
+    pdftotext -layout -f "$START" -l "$END" "$FILE" -
+  else
+    pdftotext -layout "$FILE" -
+  fi
+  EXIT_CODE=$?
+  if [ $EXIT_CODE -eq 0 ]; then
+    echo ""
+    echo "Extraction Status: SUCCESS (pdftotext)"
+    exit 0
+  fi
+  echo "pdftotext failed (exit $EXIT_CODE), trying fallback..."
+fi
+
+# Strategy 2: pandoc
+if command -v pandoc >/dev/null 2>&1; then
+  echo "Method: pandoc"
+  echo ""
+  echo "--- Content ---"
+  pandoc "$FILE" -t markdown --wrap=none 2>/dev/null
+  EXIT_CODE=$?
+  if [ $EXIT_CODE -eq 0 ]; then
+    echo ""
+    echo "Extraction Status: SUCCESS (pandoc)"
+    exit 0
+  fi
+  echo "pandoc failed (exit $EXIT_CODE), trying fallback..."
+fi
+
+# Strategy 3: PyPDF2 via Python
+if command -v python3 >/dev/null 2>&1 && python3 -c "import PyPDF2" 2>/dev/null; then
+  echo "Method: PyPDF2"
+  echo ""
+  echo "--- Content ---"
+  python3 - "$FILE" "${START:-}" "${END:-}" <<'PYEOF'
+import PyPDF2, sys
+
+path  = sys.argv[1]
+start = int(sys.argv[2]) - 1 if sys.argv[2] else 0
+end   = int(sys.argv[3])     if sys.argv[3] else None
+
+try:
+    with open(path, "rb") as f:
+        reader = PyPDF2.PdfReader(f)
+        pages  = reader.pages[start:end]
+        for i, page in enumerate(pages, start + 1):
+            print(f"\n--- Page {i} ---\n")
+            text = page.extract_text()
+            if text:
+                print(text)
+            else:
+                print("[No extractable text on this page — may require OCR]")
+    print("\nExtraction Status: SUCCESS (PyPDF2)")
+except Exception as e:
+    print(f"PyPDF2 failed: {e}", file=sys.stderr)
+    sys.exit(1)
+PYEOF
+  exit $?
+fi
+
+# All strategies failed
+echo "Extraction Status: FAILED"
+echo "No PDF extractor available."
+echo "Install options:"
+echo "  pdftotext : sudo apt install poppler-utils  (or brew install poppler)"
+echo "  pandoc    : https://pandoc.org/installing.html"
+echo "  PyPDF2    : pip install PyPDF2"
+exit 1

package/.github/skills/md-file-converter/extractors/extract-pptx.py

@@ -0,0 +1,143 @@
+#!/usr/bin/env python3
+"""
+PowerPoint extractor. Converts every slide to Markdown.
+Captures: titles, text boxes, tables, images, charts, speaker notes.
+Usage: python3 extractors/extract-pptx.py <file.pptx> [start_slide] [end_slide]
+       Slide numbers are 1-based. Omit to process all slides.
+"""
+
+import sys
+import os
+
+FILE        = sys.argv[1] if len(sys.argv) > 1 else None
+START_SLIDE = int(sys.argv[2]) if len(sys.argv) > 2 else 1
+END_SLIDE   = int(sys.argv[3]) if len(sys.argv) > 3 else None
+
+if not FILE:
+    print("Usage: extract-pptx.py <file.pptx> [start_slide] [end_slide]")
+    sys.exit(1)
+
+if not os.path.exists(FILE):
+    print(f"Error: File not found: {FILE}")
+    sys.exit(1)
+
+try:
+    from pptx import Presentation
+    from pptx.util import Inches
+    from pptx.enum.shapes import MSO_SHAPE_TYPE
+except ImportError:
+    print("Extraction Status: FAILED")
+    print("python-pptx not installed. Run: pip install python-pptx")
+    sys.exit(1)
+
+print(f"=== PowerPoint Extraction ===")
+print(f"File       : {FILE}")
+print(f"Slides     : {START_SLIDE} to {END_SLIDE or 'end'}")
+print("")
+
+prs = Presentation(FILE)
+total_slides = len(prs.slides)
+end = min(END_SLIDE, total_slides) if END_SLIDE else total_slides
+
+slide_range = range(START_SLIDE - 1, end)
+
+images_found  = 0
+tables_found  = 0
+charts_found  = 0
+notes_found   = 0
+
+for idx in slide_range:
+    slide      = prs.slides[idx]
+    slide_num  = idx + 1
+
+    print(f"---")
+    print(f"## Slide {slide_num} of {total_slides}")
+    print("")
+
+    for shape in slide.shapes:
+        # Title
+        if shape.shape_type == MSO_SHAPE_TYPE.PLACEHOLDER and hasattr(shape, "placeholder_format"):
+            ph_type = shape.placeholder_format.type
+            if ph_type is not None and ph_type == 1:  # PP_PLACEHOLDER.TITLE = 1
+                if shape.has_text_frame and shape.text_frame.text.strip():
+                    print(f"### {shape.text_frame.text.strip()}")
+                    print("")
+                continue
+
+        # Text frames
+        if shape.has_text_frame:
+            text = shape.text_frame.text.strip()
+            if text:
+                for para in shape.text_frame.paragraphs:
+                    para_text = para.text.strip()
+                    if para_text:
+                        level = para.level
+                        prefix = "  " * level + "- " if level > 0 else ""
+                        print(f"{prefix}{para_text}")
+                print("")
+
+        # Tables
+        if shape.has_table:
+            tables_found += 1
+            table = shape.table
+            print(f"**Table {tables_found}:**\n")
+            rows = table.rows
+            if rows:
+                # Header
+                header_cells = [cell.text.strip() for cell in rows[0].cells]
+                print("| " + " | ".join(header_cells) + " |")
+                print("| " + " | ".join(["---"] * len(header_cells)) + " |")
+                for row in rows[1:]:
+                    cells = [cell.text.strip() for cell in row.cells]
+                    print("| " + " | ".join(cells) + " |")
+            print("")
+
+        # Charts
+        if shape.shape_type == MSO_SHAPE_TYPE.CHART:
+            charts_found += 1
+            chart = shape.chart
+            chart_title = ""
+            try:
+                chart_title = chart.chart_title.text_frame.text if chart.has_title else "Untitled Chart"
+            except Exception:
+                chart_title = "Chart"
+
+            print(f"**Chart {charts_found}: {chart_title}**\n")
+
+            try:
+                for series in chart.series:
+                    series_name = series.name if series.name else "Series"
+                    values = list(series.values) if series.values else []
+                    print(f"- Series: {series_name}")
+                    if values:
+                        print(f"  Values: {values}")
+            except Exception:
+                print("  [Chart data not extractable — describe visually if needed]")
+            print("")
+
+        # Images
+        if shape.shape_type == MSO_SHAPE_TYPE.PICTURE:
+            images_found += 1
+            img_name = getattr(shape, "name", f"image{images_found}")
+            print(f"![{img_name}](slide{slide_num}-image{images_found}.png)")
+            print(f"> Image on slide {slide_num}. Run OCR extractor if text extraction needed.")
+            print("")
+
+    # Speaker notes
+    if slide.has_notes_slide:
+        notes_text = slide.notes_slide.notes_text_frame.text.strip()
+        if notes_text:
+            notes_found += 1
+            print(f"**Speaker Notes:**")
+            print(f"> {notes_text}")
+            print("")
+
+print("---")
+print(f"")
+print(f"=== Extraction Summary ===")
+print(f"Slides processed : {len(slide_range)} of {total_slides}")
+print(f"Tables found     : {tables_found}")
+print(f"Charts found     : {charts_found}")
+print(f"Images found     : {images_found}")
+print(f"Slides with notes: {notes_found}")
+print(f"Extraction Status: SUCCESS (python-pptx)")

package/.github/skills/md-file-converter/extractors/extract-xlsx.py

@@ -0,0 +1,175 @@
+#!/usr/bin/env python3
+"""
+Excel / ODS extractor. Converts every worksheet to Markdown tables.
+Captures: cell values, formulas, merged cells, comments, hidden sheets.
+Usage: python3 extractors/extract-xlsx.py <file> [sheet_name]
+       Omit sheet_name to process all sheets.
+"""
+
+import sys
+import os
+
+FILE       = sys.argv[1] if len(sys.argv) > 1 else None
+SHEET_NAME = sys.argv[2] if len(sys.argv) > 2 else None
+
+if not FILE:
+    print("Usage: extract-xlsx.py <file.xlsx> [sheet_name]")
+    sys.exit(1)
+
+if not os.path.exists(FILE):
+    print(f"Error: File not found: {FILE}")
+    sys.exit(1)
+
+EXT = os.path.splitext(FILE)[1].lower()
+
+# ── openpyxl (primary — XLSX/XLSM) ─────────────────────────────────────────
+def extract_openpyxl(path, target_sheet=None):
+    import openpyxl
+    from openpyxl.utils import get_column_letter
+
+    wb = openpyxl.load_workbook(path, data_only=False)  # data_only=False keeps formulas
+    wb_vals = openpyxl.load_workbook(path, data_only=True)   # data_only=True gets computed values
+
+    sheet_names = [target_sheet] if target_sheet else wb.sheetnames
+
+    for name in sheet_names:
+        if name not in wb.sheetnames:
+            print(f"## Sheet: {name}\n\n> Sheet not found.\n")
+            continue
+
+        ws      = wb[name]
+        ws_vals = wb_vals[name] if name in wb_vals.sheetnames else ws
+        hidden  = ws.sheet_state != "visible"
+
+        print(f"## Sheet: {name}{' [HIDDEN]' if hidden else ''}\n")
+
+        # Dimensions
+        if ws.max_row == 0:
+            print("> Empty sheet.\n")
+            continue
+
+        # Build table
+        rows = list(ws.iter_rows())
+        if not rows:
+            print("> Empty sheet.\n")
+            continue
+
+        # Collect all cell data
+        table_data = []
+        for row in rows:
+            row_data = []
+            for cell in row:
+                formula = str(cell.value) if cell.value is not None else ""
+                computed = ws_vals.cell(cell.row, cell.column).value
+                computed_str = str(computed) if computed is not None else ""
+
+                if formula.startswith("="):
+                    display = f"{formula} → {computed_str}"
+                else:
+                    display = formula
+
+                # Check for comment
+                if cell.comment:
+                    display += f" [Comment: {cell.comment.text.strip()}]"
+
+                row_data.append(display)
+            table_data.append(row_data)
+
+        # Render as Markdown table
+        if table_data:
+            # Header row
+            header = table_data[0]
+            col_count = max(len(r) for r in table_data)
+
+            # Pad all rows to same column count
+            for r in table_data:
+                while len(r) < col_count:
+                    r.append("")
+
+            # Print header
+            print("| " + " | ".join(header) + " |")
+            print("| " + " | ".join(["---"] * col_count) + " |")
+
+            # Print data rows
+            for row in table_data[1:]:
+                print("| " + " | ".join(row) + " |")
+
+        # Named ranges
+        if hasattr(wb, 'defined_names') and wb.defined_names:
+            named = []
+            defined_names_iter = (
+                wb.defined_names.definedName if hasattr(wb.defined_names, 'definedName')
+                else wb.defined_names.values()
+            )
+            for defn in defined_names_iter:
+                if name in str(defn.attr_text):
+                    named.append(f"  - `{defn.name}` = `{defn.attr_text}`")
+            if named:
+                print("\n**Named Ranges:**")
+                print("\n".join(named))
+
+        print("")
+
+    print("Extraction Status: SUCCESS (openpyxl)")
+
+
+# ── xlrd (fallback — older .xls) ────────────────────────────────────────────
+def extract_xlrd(path, target_sheet=None):
+    import xlrd
+    wb = xlrd.open_workbook(path)
+    sheet_names = [target_sheet] if target_sheet else wb.sheet_names()
+
+    for name in sheet_names:
+        try:
+            ws = wb.sheet_by_name(name)
+        except xlrd.XLRDError:
+            print(f"## Sheet: {name}\n\n> Sheet not found.\n")
+            continue
+
+        print(f"## Sheet: {name}\n")
+
+        if ws.nrows == 0:
+            print("> Empty sheet.\n")
+            continue
+
+        rows = [[str(ws.cell_value(r, c)) for c in range(ws.ncols)] for r in range(ws.nrows)]
+
+        print("| " + " | ".join(rows[0]) + " |")
+        print("| " + " | ".join(["---"] * ws.ncols) + " |")
+        for row in rows[1:]:
+            print("| " + " | ".join(row) + " |")
+
+        print("")
+
+    print("Extraction Status: SUCCESS (xlrd)")
+
+
+# ── Main dispatch ────────────────────────────────────────────────────────────
+print(f"=== Excel Extraction ===")
+print(f"File  : {FILE}")
+print(f"Sheet : {SHEET_NAME or 'ALL'}")
+print("")
+
+if EXT in (".xlsx", ".xlsm", ".ods"):
+    try:
+        extract_openpyxl(FILE, SHEET_NAME)
+        sys.exit(0)
+    except ImportError:
+        print("openpyxl not available, trying xlrd...")
+    except Exception as e:
+        print(f"openpyxl failed: {e}, trying xlrd...")
+
+if EXT in (".xls", ".xlsx"):
+    try:
+        extract_xlrd(FILE, SHEET_NAME)
+        sys.exit(0)
+    except ImportError:
+        print("xlrd not available.")
+    except Exception as e:
+        print(f"xlrd failed: {e}")
+
+print("Extraction Status: FAILED")
+print("No Excel library available.")
+print("Install: pip install openpyxl    (for .xlsx)")
+print("Install: pip install xlrd        (for legacy .xls)")
+sys.exit(1)

package/.github/skills/md-file-converter/scripts/check-tools.sh

@@ -0,0 +1,47 @@
+#!/usr/bin/env bash
+# Detects which extraction tools are available in the current environment.
+# Run this FIRST before any conversion. Output is used to determine strategy.
+# Usage: bash scripts/check-tools.sh
+
+echo "=== Tool Detection Report ==="
+echo ""
+
+echo "--- Document Extractors ---"
+command -v pandoc       >/dev/null 2>&1 && echo "pandoc        : AVAILABLE" || echo "pandoc        : NOT available  (DOCX/ODT/RTF fallback unavailable)"
+command -v pdftotext    >/dev/null 2>&1 && echo "pdftotext     : AVAILABLE" || echo "pdftotext     : NOT available  (PDF text extraction unavailable)"
+command -v pdfinfo      >/dev/null 2>&1 && echo "pdfinfo       : AVAILABLE" || echo "pdfinfo       : NOT available  (PDF metadata unavailable)"
+command -v pdfimages    >/dev/null 2>&1 && echo "pdfimages     : AVAILABLE" || echo "pdfimages     : NOT available  (PDF image listing unavailable)"
+command -v unrtf        >/dev/null 2>&1 && echo "unrtf         : AVAILABLE" || echo "unrtf         : NOT available  (RTF fallback unavailable)"
+
+echo ""
+echo "--- Python & Libraries ---"
+command -v python3      >/dev/null 2>&1 && echo "python3       : AVAILABLE ($(python3 --version 2>&1))" || echo "python3       : NOT available"
+
+if command -v python3 >/dev/null 2>&1; then
+  python3 -c "import openpyxl"  2>/dev/null && echo "openpyxl      : AVAILABLE" || echo "openpyxl      : NOT available  (Excel extraction unavailable — pip install openpyxl)"
+  python3 -c "from pptx import Presentation" 2>/dev/null && echo "python-pptx   : AVAILABLE" || echo "python-pptx   : NOT available  (PowerPoint extraction unavailable — pip install python-pptx)"
+  python3 -c "import PyPDF2"    2>/dev/null && echo "PyPDF2        : AVAILABLE" || echo "PyPDF2        : NOT available  (PDF fallback unavailable)"
+  python3 -c "import easyocr"   2>/dev/null && echo "easyocr       : AVAILABLE" || echo "easyocr       : NOT available  (OCR fallback unavailable — pip install easyocr)"
+fi
+
+echo ""
+echo "--- OCR Tools ---"
+command -v tesseract    >/dev/null 2>&1 && echo "tesseract     : AVAILABLE ($(tesseract --version 2>&1 | head -1))" || echo "tesseract     : NOT available  (primary OCR unavailable — install tesseract-ocr)"
+command -v exiftool     >/dev/null 2>&1 && echo "exiftool      : AVAILABLE" || echo "exiftool      : NOT available  (extended metadata unavailable)"
+
+echo ""
+echo "--- Archive Tools ---"
+command -v unzip        >/dev/null 2>&1 && echo "unzip         : AVAILABLE" || echo "unzip         : NOT available  (.zip extraction unavailable)"
+command -v tar          >/dev/null 2>&1 && echo "tar           : AVAILABLE" || echo "tar           : NOT available  (.tar/.gz extraction unavailable)"
+command -v 7z           >/dev/null 2>&1 && echo "7z            : AVAILABLE" || echo "7z            : NOT available  (.7z extraction unavailable)"
+command -v unrar        >/dev/null 2>&1 && echo "unrar         : AVAILABLE" || echo "unrar         : NOT available  (.rar extraction unavailable)"
+command -v gunzip       >/dev/null 2>&1 && echo "gunzip        : AVAILABLE" || echo "gunzip        : NOT available  (.gz extraction unavailable)"
+
+echo ""
+echo "--- Core Utilities ---"
+command -v file         >/dev/null 2>&1 && echo "file          : AVAILABLE" || echo "file          : NOT available"
+command -v stat         >/dev/null 2>&1 && echo "stat          : AVAILABLE" || echo "stat          : NOT available"
+command -v wc           >/dev/null 2>&1 && echo "wc            : AVAILABLE" || echo "wc            : NOT available"
+
+echo ""
+echo "=== End of Tool Detection ==="

package/.github/skills/md-file-converter/scripts/extract-archive.sh

@@ -0,0 +1,83 @@
+#!/usr/bin/env bash
+# Extracts any supported archive format into a target directory.
+# Usage: bash scripts/extract-archive.sh <archive_path> <output_dir>
+# After extraction, lists all files for Claude to process individually.
+
+ARCHIVE="$1"
+OUTPUT_DIR="$2"
+
+if [ -z "$ARCHIVE" ] || [ -z "$OUTPUT_DIR" ]; then
+  echo "Usage: extract-archive.sh <archive_path> <output_dir>"
+  exit 1
+fi
+
+if [ ! -f "$ARCHIVE" ]; then
+  echo "Error: Archive not found: $ARCHIVE"
+  exit 1
+fi
+
+mkdir -p "$OUTPUT_DIR"
+
+FILENAME=$(basename "$ARCHIVE")
+EXT_LOWER=$(echo "$FILENAME" | tr '[:upper:]' '[:lower:]')
+
+echo "=== Archive Extraction ==="
+echo "Archive  : $ARCHIVE"
+echo "Output   : $OUTPUT_DIR"
+echo ""
+
+# Detect format and extract
+if echo "$EXT_LOWER" | grep -q '\.zip$'; then
+  echo "Format: ZIP"
+  unzip -o "$ARCHIVE" -d "$OUTPUT_DIR" && echo "Extraction: SUCCESS" || { echo "Extraction: FAILED"; exit 1; }
+
+elif echo "$EXT_LOWER" | grep -qE '\.(tar\.gz|tgz)$'; then
+  echo "Format: TAR.GZ"
+  tar -xzf "$ARCHIVE" -C "$OUTPUT_DIR" && echo "Extraction: SUCCESS" || { echo "Extraction: FAILED"; exit 1; }
+
+elif echo "$EXT_LOWER" | grep -q '\.tar\.bz2$'; then
+  echo "Format: TAR.BZ2"
+  tar -xjf "$ARCHIVE" -C "$OUTPUT_DIR" && echo "Extraction: SUCCESS" || { echo "Extraction: FAILED"; exit 1; }
+
+elif echo "$EXT_LOWER" | grep -q '\.tar\.xz$'; then
+  echo "Format: TAR.XZ"
+  tar -xJf "$ARCHIVE" -C "$OUTPUT_DIR" && echo "Extraction: SUCCESS" || { echo "Extraction: FAILED"; exit 1; }
+
+elif echo "$EXT_LOWER" | grep -q '\.tar$'; then
+  echo "Format: TAR"
+  tar -xf "$ARCHIVE" -C "$OUTPUT_DIR" && echo "Extraction: SUCCESS" || { echo "Extraction: FAILED"; exit 1; }
+
+elif echo "$EXT_LOWER" | grep -q '\.gz$'; then
+  echo "Format: GZ (single file)"
+  DEST="$OUTPUT_DIR/$(basename "$ARCHIVE" .gz)"
+  gunzip -c "$ARCHIVE" > "$DEST" && echo "Extraction: SUCCESS" || { echo "Extraction: FAILED"; exit 1; }
+
+elif echo "$EXT_LOWER" | grep -q '\.7z$'; then
+  echo "Format: 7Z"
+  if command -v 7z >/dev/null 2>&1; then
+    7z x "$ARCHIVE" -o"$OUTPUT_DIR" -y && echo "Extraction: SUCCESS" || { echo "Extraction: FAILED"; exit 1; }
+  else
+    echo "Error: 7z not installed. Run: apt install p7zip-full"
+    exit 1
+  fi
+
+elif echo "$EXT_LOWER" | grep -q '\.rar$'; then
+  echo "Format: RAR"
+  if command -v unrar >/dev/null 2>&1; then
+    unrar x "$ARCHIVE" "$OUTPUT_DIR/" && echo "Extraction: SUCCESS" || { echo "Extraction: FAILED"; exit 1; }
+  else
+    echo "Error: unrar not installed. Run: apt install unrar"
+    exit 1
+  fi
+
+else
+  echo "Error: Unsupported archive format: $FILENAME"
+  exit 1
+fi
+
+echo ""
+echo "=== Extracted Files ==="
+find "$OUTPUT_DIR" -type f | sort
+echo ""
+echo "Total files: $(find "$OUTPUT_DIR" -type f | wc -l)"
+echo "=== End Extraction ==="

package/.github/skills/md-file-converter/scripts/extract-ocr.sh

@@ -0,0 +1,77 @@
+#!/usr/bin/env bash
+# OCR extractor for image files. Tries tesseract first, then easyocr.
+# Usage: bash scripts/extract-ocr.sh <image_path> [language]
+# language defaults to "eng". For multi-language: "eng+fra"
+
+IMAGE="$1"
+LANG="${2:-eng}"
+
+if [ -z "$IMAGE" ]; then
+  echo "Usage: extract-ocr.sh <image_path> [language]"
+  exit 1
+fi
+
+if [ ! -f "$IMAGE" ]; then
+  echo "Error: Image not found: $IMAGE"
+  exit 1
+fi
+
+echo "=== OCR Extraction ==="
+echo "Image    : $IMAGE"
+echo "Language : $LANG"
+echo ""
+
+# Strategy 1: tesseract (preferred — fast, no GPU needed)
+if command -v tesseract >/dev/null 2>&1; then
+  echo "Method: tesseract"
+  echo ""
+  echo "--- Extracted Text ---"
+  tesseract "$IMAGE" stdout -l "$LANG" 2>/dev/null
+  EXIT_CODE=$?
+  if [ $EXIT_CODE -eq 0 ]; then
+    echo ""
+    echo "OCR Status: SUCCESS (tesseract)"
+    exit 0
+  else
+    echo "tesseract failed (exit $EXIT_CODE), trying fallback..."
+  fi
+fi
+
+# Strategy 2: easyocr via Python
+if command -v python3 >/dev/null 2>&1 && python3 -c "import easyocr" 2>/dev/null; then
+  echo "Method: easyocr"
+  echo ""
+  echo "--- Extracted Text ---"
+  python3 - "$IMAGE" "$LANG" <<'PYEOF'
+import easyocr
+import sys
+
+image_path = sys.argv[1]
+lang_arg   = sys.argv[2] if len(sys.argv) > 2 else "en"
+
+# easyocr uses 2-letter codes: eng->en, fra->fr, etc.
+lang = lang_arg.split("+")[0][:2]
+
+try:
+    reader  = easyocr.Reader([lang], gpu=False, verbose=False)
+    results = reader.readtext(image_path, detail=0, paragraph=True)
+    for line in results:
+        print(line)
+    print("")
+    print("OCR Status: SUCCESS (easyocr)")
+except Exception as e:
+    print(f"easyocr failed: {e}", file=sys.stderr)
+    sys.exit(1)
+PYEOF
+  exit $?
+fi
+
+# No OCR tool available
+echo "OCR Status: FAILED"
+echo "No OCR tool available."
+echo "Install options:"
+echo "  tesseract : sudo apt install tesseract-ocr  (or brew install tesseract)"
+echo "  easyocr   : pip install easyocr"
+echo ""
+echo "Alternative: Upload the image directly into the Claude chat window to use vision-based OCR."
+exit 1

package/.github/skills/md-file-converter/scripts/get-metadata.sh

@@ -0,0 +1,57 @@
+#!/usr/bin/env bash
+# Universal metadata extractor. Tries multiple tools in priority order.
+# Usage: bash scripts/get-metadata.sh <file_path>
+# Output: key-value pairs, one per line
+
+FILE="$1"
+
+if [ -z "$FILE" ]; then
+  echo "Usage: get-metadata.sh <file_path>"
+  exit 1
+fi
+
+if [ ! -f "$FILE" ]; then
+  echo "Error: File not found: $FILE"
+  exit 1
+fi
+
+echo "=== File Metadata ==="
+echo "File Name     : $(basename "$FILE")"
+echo "File Path     : $(realpath "$FILE" 2>/dev/null || echo "$FILE")"
+echo "File Size     : $(wc -c < "$FILE" 2>/dev/null) bytes"
+echo "MIME Type     : $(file --mime-type -b "$FILE" 2>/dev/null || echo "Not Available")"
+echo "Encoding      : $(file -b "$FILE" 2>/dev/null || echo "Not Available")"
+
+# stat for timestamps (cross-platform safe subset)
+if command -v stat >/dev/null 2>&1; then
+  echo "Modified Date : $(stat -c '%y' "$FILE" 2>/dev/null || stat -f '%Sm' "$FILE" 2>/dev/null || echo "Not Available")"
+fi
+
+EXT="${FILE##*.}"
+EXT_LOWER=$(echo "$EXT" | tr '[:upper:]' '[:lower:]')
+
+# PDF-specific metadata
+if [ "$EXT_LOWER" = "pdf" ] && command -v pdfinfo >/dev/null 2>&1; then
+  echo ""
+  echo "=== PDF Metadata (pdfinfo) ==="
+  pdfinfo "$FILE" 2>/dev/null || echo "pdfinfo failed"
+fi
+
+# Extended metadata via exiftool (universal)
+if command -v exiftool >/dev/null 2>&1; then
+  echo ""
+  echo "=== Extended Metadata (exiftool) ==="
+  exiftool "$FILE" 2>/dev/null || echo "exiftool failed"
+fi
+
+# Line/word count for text files
+if file -b "$FILE" 2>/dev/null | grep -qi "text"; then
+  echo ""
+  echo "=== Text Statistics ==="
+  echo "Line Count    : $(wc -l < "$FILE" 2>/dev/null)"
+  echo "Word Count    : $(wc -w < "$FILE" 2>/dev/null)"
+  echo "Char Count    : $(wc -m < "$FILE" 2>/dev/null)"
+fi
+
+echo ""
+echo "=== End Metadata ==="