Twitter/X Account Radar 🐦

Know the moment a tracked account posts — without a single risk to a real X account.

Most X/Twitter scrapers make you re-pay for the same tweets every run, or quietly bill you for empty/error responses. This one remembers what it already showed you and never uses a logged-in account of its own — public timeline only.

Why growth/research teams use this

• Never miss a new post from accounts you watch. Point it at competitors, brands, or influencers on X; every run finds what's genuinely new since the last check.
• Pay only for what's actually new. The first run establishes a baseline of the current timeline — completely free. From then on, you're billed once per new tweet, ever, no matter how many times you re-run.
• No duplicate charges, guaranteed. Every tweet is tracked with a permanent, durable record — a re-run, a retry, or a crash never bills you twice. Found the same tweet via two different tracked accounts (e.g. a retweet)? Billed once, not twice.
• Zero account risk. No login, no cookies, no session of yours or ours — uses the same public, guest-token route your browser uses to view a public profile logged out.

How it works

1. Give it a list of X @handles you want to track and a monitorId to name this watch.
2. The first run silently catalogs the current visible timeline of each account — free, no charge, no data written yet.
3. Every run after that fetches the latest tweets and delivers only the ones published after your monitor started tracking that account — with author, text, likes/retweets/replies, and date.
4. A short digest tells you what's new across all your tracked accounts; an optional webhook can push it to Slack, Telegram, or your own system.

What you pay for

One event only — tweet-delivered. Billed once per genuinely new tweet: a reliable publish date after your monitor's baseline for that account. Your very first run's entire visible timeline is free. Repeat deliveries of a tweet you've already been billed for — impossible, by design (a permanent record per tweet prevents it, even across different tracked accounts). Errors, blocked attempts, and unreadable data cost nothing.

On the rare occasion the charge itself can't be confirmed after a tweet has already been written to your dataset, we never retry it and never bill you for it later — you keep that row for free rather than risk a duplicate charge.

What it does NOT promise

Keyword/full-text search across all of X (not just tracked accounts — this specific mode requires a much heavier technical route we haven't built yet), real-time push notifications (this is a scheduled check, not a live feed from X), sentiment analysis, retweets/likes/replies made by other users on the tweets we find (only the tracked account's own tweets), a guarantee against X changing or blocking the public route this actor uses at any time (it's not an official API), full coverage of a very high-volume account's entire history in a single baseline pass (see Limits below).

Input

Field	What it does
monitorId	Names this watch's own durable history — required, since the whole point is remembering what you've already seen
handles (1–30)	X accounts to track, with or without @
notifyOn	new_tweets (default) fires the webhook only when something new showed up; always/never for the rest
webhookUrl	HTTPS only; delivers the digest of new tweets as JSON

Output

• Dataset — one row per genuinely new tweet found: author, text, date, engagement counts, and which of your tracked accounts it was found through.
• Key-value store — a short digest (per-account counts) and a coverage report showing how many accounts answered cleanly vs. partially vs. failed, with reasons.
• Run status — a clear reason on bad input, a busy monitor, or low coverage; anything already saved and billed stays saved and billed. A run that overran its time budget is reported as failed, not silently truncated as a success.

Limits (v1)

Up to 30 accounts, up to 8 pages of tweets per account per run; one retry per account on a transient error; an early stop if too many accounts fail in a row, so a bad run doesn't burn through your budget for nothing. Your first run for a given account captures a snapshot down to that page limit — older history beyond that snapshot is not retroactively backfilled later. This never affects billing correctness: your billing baseline is fixed at the moment of that first run regardless of how deep the snapshot reached, and a tweet is never charged twice no matter what.

A note on the source

Data comes from X's public GraphQL endpoints via the same request pattern a logged-out browser makes — no login, no cookies, no account of ours or yours at risk. This is not an official API; X's Terms of Service restrict automated access, and the route itself could change or be blocked at any time — use this at your own discretion.