yapit

CLI for yapit.md — fetch clean markdown from URLs, files, and documents.

Install

uv tool install yapit

Usage

# Create document from URL, print markdown
yapit https://example.com/article
yapit https://arxiv.org/abs/2301.00001

# Local files
yapit paper.pdf
yapit notes.md

# With AI transformation (TTS annotations, spoken readings for math, figure extraction, noise filtering)
yapit paper.pdf --ai

# Fetch existing document
yapit https://yapit.md/listen/<doc-id>
yapit <doc-id> --annotated

# Save to directory with images
yapit <doc-id> -o .

# Stdin
echo "hello world" | yapit -

Auth

Creating documents or accessing private docs requires a yapit.md account and an API token. Create one in your account settings, then:

export YAPIT_TOKEN=yapit_...

The CLI sends it as Authorization: Bearer. Fetching shared documents works without auth.

Self-hosting

Point the CLI at your own instance with YAPIT_BASE_URL:

export YAPIT_BASE_URL=http://localhost
yapit fetch paper.pdf

YAPIT_BASE_URL is the address you open yapit at in the browser; the CLI calls the API under its /api path.

If your server runs with AUTH_ENABLED=false (the make self-host default), no token is needed. If auth is enabled on your instance, set YAPIT_TOKEN as above. The self-hosting guide in yapit's docs/auth.md shows how to create a token on the server.

Save to directory (-o)

-o <dir> saves markdown, TTS annotations, and images:

<dir>/<slug>/
  <slug>.md       # clean markdown, image paths rewritten to relative
  TTS.md          # annotated version (yap-speak, yap-show tags)
  <slug>-<N>.png  # images renamed sequentially

Errors if the output directory already exists.