file-conversion
/file-conversion
Konwertuje pliki między formatami: obrazy JPG, PNG, WEBP, HEIC i SVG, klipy wideo oraz dokumenty DOCX, PPTX, Markdown i PDF.
- Autor
- aomni-com ↗
- Repozytorium
- open-duet ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 30.10.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 4
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Brak sygnałów. Repozytorium zmienione 30.10.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Konwertuje pliki między formatami: obrazy JPG, PNG, WEBP, HEIC i SVG, klipy wideo oraz dokumenty DOCX, PPTX, Markdown i PDF.
Oryginalny opis (EN)
Convert files between formats — images (JPG/PNG/WEBP/HEIC/SVG via Pillow, pillow-heif, cairosvg), video (clip extraction and video-to-GIF via moviepy/ffmpeg), and documents (DOCX/PPTX ↔ Markdown ↔ PDF via pypandoc, python-docx, python-pptx, libreoffice, markitdown). Use when the user asks to convert, transform, resize, rasterize, re-encode, or change the format of an image, video clip, or document file.
Jak działa
- Pick the right tool
- When to load each reference
- Defaults
- Minimal example — image format conversion
- Dependencies
- Verify before reporting done
Podgląd SKILL.md
---
name: file-conversion
target: both
description: Convert files between formats — images (JPG/PNG/WEBP/HEIC/SVG via Pillow, pillow-heif, cairosvg), video (clip extraction and video-to-GIF via moviepy/ffmpeg), and documents (DOCX/PPTX ↔ Markdown ↔ PDF via pypandoc, python-docx, python-pptx, libreoffice, markitdown). Use when the user asks to convert, transform, resize, rasterize, re-encode, or change the format of an image, video clip, or document file.
---
# File Conversion
Convert image, video, and document files between formats. Pick the library based on the conversion type, then load the matching reference for code patterns.
## Pick the right tool
| Need | Use | Reference |
| --------------------------------- | -------------------- | ----------------------------------- |
| JPG ↔ PNG ↔ WEBP | Pillow | `references/image-conversion.md` |
| HEIC → anything | pillow-heif + Pillow | `references/image-conversion.md` |
| SVG → raster | cairosvg | `references/image-conversion.md` |
| Video → GIF, clip extraction | moviepy | `references/video-to-gif.md` |
| DOCX/PPTX/Markdown/PDF interconv. | pypandoc + helpers | `references/document-conversion.md` |
For raster → SVG: not supported. SVG can only be rasterized, not generated from a bitmap.
## When to load each reference
- **`references/image-conversion.md`** — load when converting any image. Includes the RGBA→RGB JPEG handling, HEIC registration, and SVG rasterization patterns.
- **`references/video-to-gif.md`** — load when extracting a clip from a video, generating a GIF, or optimizing a GIF for web (size targets, FPS, dimensions).
- **`references/document-conversion.md`** — load when touching DOCX, PPTX, Markdown, or PDF as a document (not as a generated artifact — for PDF generation/extraction use the `pdf` skill).
Do not load all three at once. Load only the one that matches the conversion direction.
## Defaults
- Images: write JPEG with `quality=95, optimize=True`. Convert RGBA → RGB before writing JPEG.
- GIFs: target ≤ 10 MB. Start at width 480, FPS 12. Drop FPS first if over budget, then width.
- Documents: prefer `pypandoc` for DOCX ↔ Markdown ↔ PDF. Use `python-pptx` for PPTX-specific work.
## Minimal example — image format conversion
```python
from PIL import Image
img = Image.open("input.png")
if img.mode == "RGBA":
img = img.convert("RGB")
img.save("output.jpg", quality=95, optimize=True)
```
For HEIC, register `pillow-heif` first; for SVG input, use `cairosvg` to rasterize. See `references/image-conversion.md`.
## Dependencies
System packages are pre-installed (ffmpeg, libvips-tools, pandoc, poppler-utils). Python packages are installed on demand:
```bash
pip install Pillow pillow-heif cairosvg moviepy python-docx python-pptx pypandoc
```
If a conversion requires LaTeX (some pypandoc PDF paths), prefer the libreoffice route instead — LaTeX is not installed by default.
## Verify before reporting done
After every conversion, confirm:
1. The output file exists and is non-empty.
2. The output dimensions / page count / duration match what was asked.
3. For lossy formats, the result is visually acceptable — not just byte-valid.
Pliki w paczce (4)
- SKILL.md 3 KB
- references/document-conversion.md 15 KB
- references/image-conversion.md 12 KB
- references/video-to-gif.md 12 KB
Wersje
| Repozytorium | Wynik | Licencja |
|---|---|---|
| Ta strona · aomni-com/open-duet | 100 | MIT |
| wshobson/agents | 100 | MIT |
| aadilr/changethisfile-mcp | 100 | MIT |
Migawka bazy z 13.11.2026 · wersja paczki 9020fed4e99e (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 100convert-image-to-jpg
Konwertuje lokalne obrazy z formatów takich jak HEIC, PNG, TIFF, WebP, BMP, GIF i AVIF do JPG z ustawionym docelowym DPI (domyślnie 96). - 100markitdown
Konwertuje dokumenty PDF, PPTX, DOCX, XLSX i starsze formaty Office do Markdown przez narzędzia MCP MarkItDown, na potrzeby czytania i wyciągania tekstu. - 100converting-files
Konwertuje pliki między formatami w kontenerze (dokumenty, obrazy, audio, wideo), dobierając do pary formatów silnik: pandoc, LibreOffice, ImageMagick lub ffmpeg. - 100file-converter
Konwertuje i przekształca pliki: obrazy (rozmiar, format, HEIC), markdown do PDF i HTML, dane CSV, JSON, YAML, TOML i XML, SVG, base64 i kodowanie tekstu. - 100filestar
Konwertuje, zmienia rozmiar, kompresuje i łączy pliki lokalnie przez Filestar, także PDF, OCR i pliki CAD. - 100compdf-conversion-cli
Konwertuje PDF i obrazy do 10 formatów (Word, Excel, PPT, HTML, obraz, TXT, JSON, Markdown, RTF, CSV) z analizą układu AI i OCR.