WebConverter

Ask an agent to

• Convert my DOCX to PDF
• Merge these three PDFs into one
• Extract the audio from this video as MP3

15 tools

• convert_3d_to_gltfread-only

  Convert a 3D scene (FBX, OBJ, DAE, PLY, STL, GLTF, GLB, 3DS, LWO, BLEND, X, MS3D, ASE) to glTF 2.0 or binary GLB, entirely in the browser. Native magnum-plugins importers handle the mainstream formats (ufbx for…

  fileBase64: string, fileName: string?, outputFormat: string?, optimize: boolean?, simplifyRatio: number?

• convert_audioread-only

  Convert an audio file — or extract the audio track from a video — to MP3, OGG, WAV or FLAC. Decoding uses the browser's codecs, so anything the browser can play is accepted.

  fileBase64: string, fileName: string?, targetFormat: string

• convert_documentread-only

  Convert a document between formats with Pandoc (WebAssembly). Common inputs: docx, odt, rtf, html, markdown, latex, rst, mediawiki, epub. Common outputs: markdown, html, plain, latex, rst, asciidoc, docx, odt. The…

  fileBase64: string, fileName: string?, fromFormat: string, toFormat: string

• convert_imageread-only

  Convert an image to another format entirely in the browser — no upload. WASM importers accept BMP, DDS, GIF, HDR, ICO, JPEG, KTX, PGM, PIC, PNG, PPM, PSD, TGA, WebP; a browser-decode fallback adds HEIC (Safari/iOS)…

  fileBase64: string, fileName: string?, targetFormat: string

• convert_videoread-only

  Transcode a video to MP4 (H.264 + AAC), WebM (VP9 + Opus) or animated GIF using ffmpeg-wasm in the browser. The first call lazily downloads the ffmpeg core for the chosen output (~30 MB per variant); cached afterwards.

  fileBase64: string, fileName: string?, targetFormat: string, config: object?

• delete_pdf_pagesread-only

  Drop the listed zero-based page indices from a PDF, writing a new PDF of the remaining pages, in their original order.

  pdfBase64: string, pagesToDelete: array, fileName: string?

• extract_pdf_textread-only

  Pull text out of a PDF (pdf.js) — plain text by default, or simple Markdown with one `## Page N` heading per page when `asMarkdown: true`.

  pdfBase64: string, asMarkdown: boolean?, fileName: string?

• images_to_pdfread-only

  Combine one or more images into a single PDF, one image per page, in order, in the browser.

  imagesBase64: array, fileName: string?

• images_to_searchable_pdfread-only

  Combine images into a PDF *and* run OCR (Tesseract.js, lazy-loaded) so the resulting PDF has selectable, searchable text on top of the original pixels. Useful for scans and screenshots.

  imagesBase64: array, fileName: string?, ocrLanguage: string?

• list_supported_formatsread-only

  Enumerate every format and tool WebConverter exposes via WebMCP.

• merge_pdfsread-only

  Merge multiple PDF files into a single PDF (all pages, in order) — in the browser, no upload.

  pdfsBase64: array, fileName: string?

• remove_image_backgroundread-only

  Produce a version of the image with the background made transparent (or a chosen colour). Uses a tiny U²-Net-P ONNX model (≈4 MB) for the initial mask, then a deterministic WASM matting pass to clean the edges. Output…

  fileBase64: string, fileName: string?, outputFormat: string?

• reorder_pdf_pagesread-only

  Reorder the pages of a single PDF and write a new PDF with the given order. pageOrder is an array of zero-based source-page indices, in the desired order.

  pdfBase64: string, pageOrder: array, fileName: string?

• transcribe_audioread-only

  Transcribe an audio file — or the audio track of a video — to text with timestamps using a quantised Whisper model (whisper.cpp compiled to WASM). Runs entirely in the browser. The first call downloads the model (~40…

  fileBase64: string, fileName: string?

• trim_videoread-only

  Cut a clip from startTime → endTime (seconds) out of a video using ffmpeg-wasm, without re-uploading anything. Output is MP4 by default.

  fileBase64: string, fileName: string?, startTime: number, endTime: number, outputExt: string?