Formats

One read call handles every format below, detected from the file's bytes, not its name. Everything is parsed natively in Rust (zip/XML, calamine, html5ever): no Python, LibreOffice, or cloud service.

Input	What you get
PDF	Reading-order Markdown with headings, lists, tables, and page anchors
Word .docx	Headings, formatting, links, lists, tables, footnotes, equations
PowerPoint .pptx	One section per slide, with notes and chart data
Excel .xlsx .xls .ods · CSV/TSV	One table per sheet
EPUB	One section per chapter
HTML and URLs	The main article only
Markdown, text, JSON	Unchanged, with pagination
Images	Metadata, EXIF, and OCR text
Audio / video	Metadata, chapters, subtitles, and transcripts

PDF

Reading-order Markdown: headings, paragraphs, lists, tables, sub/superscripts, <!-- page N --> markers, and bookmarks as an outline. Running headers and page numbers are removed. Image-only pages are OCR'd when tesseract is installed.

How it works: anymd reads glyph positions rather than text runs. Glyphs are grouped into lines by baseline, which tolerates super- and subscripts. Word spaces come from the gaps between glyphs, measured against the font size and adjusted for letter tracking. A column-aware XY cut finds gutters between running text. Rows whose cells line up become pipe tables. Pages are processed in parallel and isolated from each other, so one malformed page never fails the whole document.

pages selects PDF pages. For images, geometry, or JSON structure, use inspect.

Word

Headings, bold/italic, links, nested lists, tables with merged cells, footnotes, and equations as LaTeX.

PowerPoint

One section per slide in deck order: titles, bullets, tables, chart data, and speaker notes. pages selects slides.

Spreadsheets

.xlsx, .xls, .ods, CSV, and TSV. One Markdown table per sheet, dates as ISO strings, capped at 2,000 rows per sheet. pages selects sheets.

EPUB

One section per chapter in spine order, plus title and author. pages selects chapters.

HTML and URLs

The main article only: navigation, cookie banners, and sidebars are dropped. Relative links are resolved, and code blocks keep their language. URL fetches are guarded; see Security.

Markdown, text, JSON

Returned unchanged, with pagination and the token budget.

Images

Dimensions and EXIF (camera, date, GPS), plus OCR text when tesseract is installed.

Audio and video

Duration, streams, chapters, and embedded and sidecar subtitles (SRT/VTT), via ffprobe/ffmpeg. With transcript: true (CLI: --transcript), a local whisper.cpp transcript.

Transcripts

A transcript needs three things on your machine; nothing is uploaded.

• whisper.cpp: whisper-cli (or whisper-cpp) on PATH, or ANYMD_WHISPER_BIN pointing at it. macOS: brew install whisper-cpp. Windows: whisper-bin-x64.zip from the whisper.cpp releases. Linux: build from source (git clone https://github.com/ggml-org/whisper.cpp && cd whisper.cpp && cmake -B build && cmake --build build -j --config Release, then put build/bin/whisper-cli on PATH), the release tarball, or Homebrew.
• ffmpeg, to extract the audio track.
• A ggml model. ANYMD_WHISPER_MODEL wins when set; otherwise anymd uses a ggml-*.bin in its cache ($ANYMD_CACHE_DIR/models, else ~/.cache/anymd/models, ~/Library/Caches/anymd/models, or %LOCALAPPDATA%\anymd\cache\models). To fetch one on first use, pass download_whisper_model: true (CLI: --download-whisper-model, which implies --transcript) or set ANYMD_WHISPER_AUTO_DOWNLOAD=1. anymd downloads ggml-base.en.bin (148 MB) from the official ggerganov/whisper.cpp repository, checks its SHA-256, and renames it into place. ANYMD_WHISPER_MODEL_SIZE picks tiny, tiny.en, base, base.en, small, or small.en; ANYMD_WHISPER_MODEL_BASE_URL points at a mirror (the hash is still checked). Multilingual models detect the spoken language.

anymd talk.mp4 --download-whisper-model

When a piece is missing, the output names it with the install command for your OS instead of failing.

Optional tools

anymd never needs these, but uses them when they are on your PATH:

Tool	Adds
tesseract	OCR for images and scanned PDF pages
ffprobe	Audio/video metadata and chapters
ffmpeg	Embedded subtitles and transcript audio
whisper-cli (whisper.cpp)	Local transcripts (see Transcripts)

Check what anymd found:

$ anymd doctor
anymd 6.0.0 (native Rust)
  tesseract    found      OCR for images and scanned PDF pages
  ffprobe      found      audio/video metadata and chapters
  ffmpeg       found      embedded subtitles and transcript audio
Transcripts (--transcript):
  whisper.cpp    found      /opt/homebrew/bin/whisper-cli
  whisper model  not found  cache /Users/me/Library/Caches/anymd/models; `--download-whisper-model` fetches ggml-base.en.bin (148 MB)

Typical installs: brew install tesseract ffmpeg whisper-cpp on macOS, apt install tesseract-ocr ffmpeg on Debian/Ubuntu. For other OCR languages, install the tesseract language pack (for example tesseract-ocr-chi-tra).