POST /v1/scrape — scrape a URL and return content in Markdown, screenshot, PDF, CSV, or HTML.
Scrape a URL
POST /v1/scrape fetches a URL with a stealth browser and returns its content in one
of five output formats. Markdown is the default and is optimized for LLM consumption.
Authentication: API key or x402 payment. On x402 requests, payment is verified
before the scrape and settled only on success.
Endpoint
POST https://api.tazpal.com/v1/scrape
Request
Headers
Header
Required
Description
Authorization: Bearer baas_live_...
one of
API key authentication
X-PAYMENT: <base64 permit>
one of
x402 payment
Content-Type: application/json
yes
Must be application/json
X-Request-ID
no
Client-supplied request ID (echoed in X-Request-ID response header)
Body
Field
Type
Required
Default
Description
url
string
✅
—
Target URL to scrape
output
string
no
markdown
One of markdown, screenshot, pdf, csv, html
options
object
no
{}
Output-specific options (see below)
wait_for_selector
string
no
—
CSS selector to wait for before extraction
timeout_ms
integer
no
20000
Navigation timeout in ms (1000–120000)
block_media
boolean
no
true
Block image/font/video requests for speed
proxy_url
string
no
—
Explicit proxy (socks5:// or http://)
wait_strategy
string
no
auto
default, spa, heavy, or cloudflare
retry
boolean
no
true
Retry on failure with exponential backoff
bypass_cache
boolean
no
false
Skip the 5-minute response cache
javascript
string
no
—
Custom JS to execute after page load
Options by output format
screenshot options
Key
Type
Default
Description
format
string
png
png or jpeg
full_page
boolean
true
Capture the full page (PNG only)
width / height
integer
—
Explicit viewport size
quality
integer
90
JPEG quality
pdf options
Key
Type
Description
format
string
Page size (e.g. A4, Letter)
landscape
boolean
Landscape orientation
print_background
boolean
Print background graphics
scale
number
Render scale (e.g. 0.8)
margin
object
{top, bottom, left, right} margins
quality
integer
JPEG quality for the screenshot fallback
csv options
Key
Type
Default
Description
table_selector
string
table
CSS selector for tables to extract
Response
Success — 200 OK
Field
Type
Description
status
integer
Always 200
url
string
The scraped URL
output
string
The output format used
data
object
ScrapeData for markdown; format-specific dict otherwise