X / Twitter Mentions & Keyword Monitor

Monitor X (Twitter) for mentions of the handles you care about and for a watchlist of keywords or phrases. Every matching public post is returned as a clean, flat record with full engagement metadata — ready for brand monitoring, competitive intelligence, lead signals, and social listening.

Point it at a set of handles and keywords, schedule it to run on an interval, and build a continuously updated feed of who is talking about your brand, products, competitors, or topics.

What it does

• Mention monitoring — track public posts that mention one or more accounts.
• Keyword & phrase watchlists — track posts matching your terms; multi-word entries are matched as exact phrases.
• Advanced queries — power users can supply a raw search query with operators such as from:, to:, filter:, and OR.
• Incremental runs — a rolling "since minutes" window plus an automatic per-monitor checkpoint means scheduled runs pick up only what is new.
• Normalized output — consistent fields and engagement counts across every result, regardless of which monitor matched it.

Input

Field	Type	Description
handles	array	Accounts to watch for mentions. Accepts @handle, bare handle, or profile URL.
keywords	array	Keywords or phrases to watch. Multi-word entries match as an exact phrase.
query	string	Optional raw advanced search query for power users.
sort	string	latest (newest first, best for monitoring) or top (most relevant).
maxItems	integer	Maximum posts to collect per run (1–2000).
since	string	Optional ISO date (YYYY-MM-DD); skip posts before it.
until	string	Optional ISO date (YYYY-MM-DD); skip posts on/after it.
sinceMinutes	integer	Optional rolling window in minutes; ideal for scheduled runs.
includeReplies	boolean	Include posts that are replies. Default true.
includeRetweets	boolean	Include reposts (retweets). Default false.
language	string	Optional ISO language code (e.g. en, es).
authCookie	string (secret)	Optional — your own logged-in X session cookie.
authHeader	string (secret)	Optional — custom Authorization header.
bearerToken	string (secret)	Optional — bearer token override.
proxyConfiguration	object	Proxy settings; residential proxies recommended.

Provide at least one of handles, keywords, or query.

Example

{
  "handles": ["@apify"],
  "keywords": ["web scraping", "data extraction"],
  "sort": "latest",
  "maxItems": 100,
  "sinceMinutes": 60,
  "includeReplies": true,
  "includeRetweets": false,
  "language": "en",
  "proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

Output

Each dataset item is one matched post:

Field	Description
monitorType / monitorValue	Which monitor matched (mention / keyword / query) and its value.
tweetId / tweetUrl	Post ID and direct URL.
text	Full post text.
authorUsername / authorName / authorId	Author identity.
createdAt / language	Creation timestamp (ISO 8601 UTC) and language.
replyCount / repostCount / likeCount / quoteCount / viewCount	Engagement metrics.
isReply / isRepost / inReplyToTweetId	Conversation context.
mentionedUsers / hashtags / links	Entities extracted from the post.
fetchedAt	When the record was collected (ISO 8601 UTC).

Important notes

• Authentication. Searching X reliably generally requires valid authenticated session details together with reliable proxy access. The actor accepts optional, securely-stored session fields you provide. Without them it makes a best-effort attempt and, if access is declined, finishes with a clear message and no invented data — you are not charged for empty results.
• Scheduling. Combine sinceMinutes with Apify's scheduler to keep a rolling, deduplicated feed up to date. The actor remembers the newest post it has seen for each monitor between runs.
• Responsible use. Only collect public data and respect X's Terms of Service and applicable laws and rate limits.