Web change monitoring

POST

scrape

curl --request POST \
  --url https://apix.pline.ai/v1/scrape \
  --header 'X-API-Key: <api-key>'

const options = {method: 'POST', headers: {'X-API-Key': '<api-key>'}};

fetch('https://apix.pline.ai/v1/scrape', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://apix.pline.ai/v1/scrape"

	req, _ := http.NewRequest("POST", url, nil)

	req.Header.Add("X-API-Key", "<api-key>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

Watch a page for meaningful changes — a pricing page, a terms-of-service update, a competitor’s landing page — without storing and diffing raw HTML by hand.

Use case

Alert when a competitor changes pricing, copy, or product availability
Track terms-of-service, policy, or changelog pages for updates
Detect when a page’s content stops matching what you last indexed

Step 1: Take a clean snapshot

Request markdown instead of raw html — boilerplate like ads, session tokens, or rotating banners get stripped by default (only_main_content: true), so the diff only reflects real content changes. The page text comes back in data.markdown.

curl -X POST "https://apix.pline.ai/v1/scrape" \
  -H "X-API-Key: $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://competitor.example.com/pricing",
    "output": ["markdown"],
    "tag": "monitor.pricing-page"
  }' | jq -r '.data.markdown'

import os
import httpx

response = httpx.post(
    "https://apix.pline.ai/v1/scrape",
    headers={"X-API-Key": os.environ["API_KEY"]},
    json={
        "url": "https://competitor.example.com/pricing",
        "output": ["markdown"],
        "tag": "monitor.pricing-page",
    },
    timeout=60.0,
)
response.raise_for_status()
markdown = response.json()["data"]["markdown"]

const response = await fetch("https://apix.pline.ai/v1/scrape", {
  method: "POST",
  headers: {
    "X-API-Key": process.env.API_KEY,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    url: "https://competitor.example.com/pricing",
    output: ["markdown"],
    tag: "monitor.pricing-page",
  }),
});

const { data } = await response.json();
const markdown = data.markdown;

Step 2: Detect a change

Hash the Markdown and compare it with the hash from the last run. This script is complete: save it as monitor.py, set API_KEY, and run it. It keeps the last hash for each URL in a local JSON file, so the first run saves a baseline and later runs report changes.

Python

import hashlib
import json
import os
from pathlib import Path

import httpx

URL = "https://competitor.example.com/pricing"
STATE_FILE = Path("monitor_state.json")


def fetch_markdown(url: str) -> str:
    response = httpx.post(
        "https://apix.pline.ai/v1/scrape",
        headers={"X-API-Key": os.environ["API_KEY"]},
        json={"url": url, "output": ["markdown"], "tag": "monitor.pricing-page"},
        timeout=60.0,
    )
    response.raise_for_status()
    return response.json()["data"]["markdown"]


state = json.loads(STATE_FILE.read_text()) if STATE_FILE.exists() else {}

markdown = fetch_markdown(URL)
content_hash = hashlib.sha256(markdown.encode()).hexdigest()

if URL not in state:
    print("Baseline saved.")
elif state[URL] != content_hash:
    print(f"CHANGED: {URL}")  # replace with a Slack message, email, or webhook call
else:
    print("No change.")

state[URL] = content_hash
STATE_FILE.write_text(json.dumps(state, indent=2))

To see what changed, also save the Markdown from each run and compare the two versions with difflib.unified_diff.

Step 3: Run it on a schedule

Run monitor.py from a cron job, a queue worker, or a scheduled function — hourly for fast-moving pages, daily for slower ones. For example, hourly with cron:

0 * * * * cd /path/to/monitor && API_KEY=your-api-key python3 monitor.py

Use it with a database instead of the JSON file once you monitor many pages. Use a distinct tag per monitored page so you can audit run history and cost per target in Request history.

Request highlights

Setting	Value	Why
`output`	`["markdown"]`	Converts the page to text, filtering out most incidental markup noise
`only_main_content`	`true` (default)	Strips navigation/footers so they don’t trigger false-positive diffs
`tag`	per monitored URL	Lets you trace and cost-attribute each page’s monitoring history

Tips

Diff at the paragraph or section level instead of full-text equality if the page has any dynamic content (timestamps, view counts) you want to ignore.
For JS-heavy pages, set js_render: true and a wait_selector so you snapshot the rendered state, not a loading skeleton — see Scrape.
If you need structured fields (e.g. just the price, not the whole page), add prompt/schema with output: ["json"] instead of diffing raw text — see Output.

Web change monitoring

curl --request POST \
  --url https://apix.pline.ai/v1/scrape \
  --header 'X-API-Key: <api-key>'

const options = {method: 'POST', headers: {'X-API-Key': '<api-key>'}};

fetch('https://apix.pline.ai/v1/scrape', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://apix.pline.ai/v1/scrape"

	req, _ := http.NewRequest("POST", url, nil)

	req.Header.Add("X-API-Key", "<api-key>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

​Use case

​Step 1: Take a clean snapshot

​Step 2: Detect a change

​Step 3: Run it on a schedule

​Request highlights

​Tips

​Related

Use case

Step 1: Take a clean snapshot

Step 2: Detect a change

Step 3: Run it on a schedule

Request highlights

Tips

Related