ext-page-monitor

name

ext-page-monitor

description

Watch web pages for changes on demand — competitor, price and promo monitoring. Fetches a page, extracts readable text (markup stripped), stores a timestamped snapshot, and produces a unified diff against the previous snapshot. Pure standard library, no MCP, no pip deps, runs fully offline. Use when the user wants to track changes on a competitor page, detect price/promo updates, snapshot a page, or diff a page against the last time it was checked.

ext-page-monitor — on-demand page change monitor

Self-contained BDOS extension that watches web pages for content changes. It fetches a page with the Python standard library only (urllib), extracts readable text (scripts/styles/markup stripped, whitespace collapsed), stores a timestamped snapshot on disk, and diffs new fetches against the previous one.

No MCP server, no pip dependencies, no venv — works offline and survives bdos update because it lives under my/. Snapshots are stored under page_monitor/snapshots/ (gitignored — never committed).

Use this to answer "did this page change since last time?", "did the competitor drop a promo?", or "did the price move?" — run it on demand whenever you want a check.

Language

Talk to the user in their language (PL or EN — match how they wrote to you). Code, logs and saved files stay in English.

Result shape

Every function returns a dict with an ok key. On network error: {"ok": False, "error": "..."}.

• snapshot(url) → ok, url, hash, path, changed_vs_previous (bool | None)
• diff(url) → ok, url, changed, added_lines, removed_lines, diff, path (plus note when a baseline is first created)
• list_snapshots(url) → ok, url, count, snapshots (list of {fetched_at, hash, path})

Take a snapshot

Fetch the page and store it. changed_vs_previous is None on the very first snapshot, otherwise a bool vs the most recent prior snapshot.

import sys; sys.stdout.reconfigure(encoding='utf-8')
from my.extensions.page_monitor import snapshot
r = snapshot("https://competitor.com/pricing")
if r["ok"]:
    print("hash:", r["hash"])
    print("changed vs previous:", r["changed_vs_previous"])
    print("saved:", r["path"])
else:
    print("ERROR:", r["error"])

Diff a page against the previous snapshot

Takes a fresh snapshot, then produces a unified diff against the previous one. The first call just creates a baseline (changed=False, see note).

import sys; sys.stdout.reconfigure(encoding='utf-8')
from my.extensions.page_monitor import diff
r = diff("https://competitor.com/pricing")
if r["ok"]:
    if r.get("note"):
        print(r["note"])
    else:
        print(f"changed: {r['changed']} | +{r['added_lines']} -{r['removed_lines']}")
        print(r["diff"])          # unified diff, truncated to ~8000 chars
else:
    print("ERROR:", r["error"])

List stored snapshots for a URL

import sys; sys.stdout.reconfigure(encoding='utf-8')
from my.extensions.page_monitor import list_snapshots
r = list_snapshots("https://competitor.com/pricing")
print("count:", r["count"])
for s in r["snapshots"]:
    print(s["fetched_at"], s["hash"][:12], s["path"])

Notes

• On demand only. There is no cron or scheduler — call these when you want a check. To watch a page over time, snapshot/diff it periodically yourself.
• Content, not markup. Diffs are computed on extracted readable text, so layout/markup churn does not create false positives.
• Custom timeout. All functions accept timeout=<seconds> (default 20).
• Snapshots are local. They live under page_monitor/snapshots/<domain>/ and are gitignored — do not commit them.
• Combine with BDOS data. E.g. diff a competitor's pricing page, then relate observed price/promo changes to your own campaign or Merchant Center data.