Skip to content
LoginGet started

Media tools your agent already has

Someone sends a photo, a voice note or a video. Your agent can open it, change it, and send back what you asked for.

41 tools

Media arrives in messaging apps constantly, and an agent without these tools can only say it received something. With them it converts the format your phone produced into the one you needed, pulls a frame out of a clip, strips audio to transcribe it, or resizes a photo before it goes anywhere.

The tools

Preinstalled on every plan, with nothing to set up. A few depend on the engine you pick.

  • ffmpeg

    Convert, trim, resize and re-encode audio and video.

  • ffprobe

    Read codecs, streams, duration and metadata from a media file.

  • magick

    ImageMagick. Resize, crop, convert and compose images.

  • convert

    The ImageMagick classic: resize, crop and convert in one call.

  • img2pdf

    Wrap images into a PDF without re-encoding them.

  • rsvg-convert

    Render SVG to PNG, PDF or PostScript at any size.

  • pdftoppm

    Render PDF pages to PNG, JPEG or PPM images.

  • exiftool

    Read, write and strip metadata in photos and videos.

  • pandoc

    Convert between Markdown, HTML, DOCX, LaTeX and more.

  • soffice

    LibreOffice, headless: convert documents, sheets and slides.

  • pdftotext

    Extract the text layer out of a PDF.

  • qpdf

    Split, merge, rotate and repair PDF files.

  • gs

    Ghostscript. Compress, flatten and convert PDF and PostScript.

  • tesseract

    OCR: pull text out of scans, photos and screenshots.

  • dot

    Graphviz. Render graphs and diagrams from a text description.

  • identify

    Report an image’s format, dimensions, color space and EXIF data.

  • mogrify

    Resize, convert or strip a whole batch of images in place.

  • compare

    Diff two images and write out a visual difference map.

  • montage

    Tile images into a contact sheet with labels and spacing.

  • composite

    Overlay one image on another with a mask, an offset or a blend mode.

  • pdfinfo

    Print a PDF’s page count, page size, metadata and encryption state.

  • pdfimages

    Extract every embedded image from a PDF at its original resolution.

  • pdftocairo

    Render PDF pages to PNG, JPEG, SVG or PostScript through Cairo.

  • pdftohtml

    Convert a PDF to HTML with its layout and images preserved.

  • pdftops

    Convert a PDF to PostScript for printing or further processing.

  • pdffonts

    List the fonts a PDF uses and whether each one is embedded.

  • pdfseparate

    Split a PDF into one file per page.

  • pdfunite

    Merge several PDFs into one, in the order you list them.

  • pdfsig

    Verify the digital signatures on a signed PDF.

  • ps2pdf

    Convert PostScript or EPS to PDF through Ghostscript.

  • pdf2ps

    Convert a PDF back to PostScript through Ghostscript.

  • lowriter

    Convert documents between PDF, DOCX and ODT without a display.

  • localc

    Convert spreadsheets between XLSX, ODS and CSV headlessly.

  • loimpress

    Convert slide decks between PPTX, ODP and PDF headlessly.

  • neato

    Lay out an undirected graph with a spring model instead of dot’s ranks.

  • edge-tts

    Synthesize speech from text and write it to an audio file.

  • elevenlabs

    Synthesize speech with ElevenLabs voices and write it to a file.

  • vips

    Convert and transform images with a fast, low-memory pipeline.

  • vipsthumbnail

    Generate thumbnails at catalogue scale without loading whole images.

  • avifenc

    Encode images to AVIF for smaller pages and faster loads.

  • oxipng

    Shrink PNG files losslessly, without touching how they look.

Get started

A real command line, already set up

Every agent runs on a sandboxed machine with the whole toolbox preinstalled. No setup step, on any plan.

Build your agentTalk to us