Skip to main content
The power of Sorsa API lies in its ability to handle Advanced Search Operators. While most scrapers only search for simple keywords, Sorsa allows you to build surgical queries to filter through billions of tweets. Whether you are doing OSINT research, brand sentiment analysis, or competitive benchmarking, mastering these operators is essential for getting high-quality data.

💡 Use Cases for Advanced Filtering

  • Brand Monitoring: Find every mention of your product while excluding your own official accounts.
  • Competitor Audits: Extract only the most viral content from a competitor using min_faves.
  • Lead Generation: Find users asking questions in your niche by combining keywords with the ? operator.
  • Media Archiving: Collect all high-resolution images or videos from a specific event or news story.

🛠 How to Build Your Query

There are three ways to construct your search string for the /search-tweets endpoint:
  1. Sorsa Search Builder: Use our Visual Playground. It’s a GUI where you toggle filters, and it generates the API-ready string for you.
  2. Manual Construction: Use the operators listed below to build a string (e.g., (Marketing OR Advertising) min_faves:100).
  3. X Native Copy-Paste: Go to X Advanced Search, apply your filters, hit search, and simply copy the text from the search bar directly into your API request.

📖 The Complete Operator Dictionary

1. Basic Logic & Keywords

OperatorLogicWhat it finds…
nasa esaANDTweets containing both “nasa” and “esa”.
nasa OR esaORTweets containing either “nasa” or “esa”.
"state of the art"PhraseMatches the exact phrase.
-keywordNOTExcludes tweets containing the word.
(design OR "user experience")GroupGroups operators for complex logic.
+radioooooForceForces exact match (prevents spelling correction).

2. Engagement & Quality Control

  • min_faves:X — Tweets with at least X likes.
  • min_retweets:X — Tweets with at least X retweets.
  • min_replies:X — Tweets with at least X replies.
  • filter:verified — Only tweets from “Legacy” verified accounts.
  • filter:blue_verified — Only tweets from X Premium (Blue) subscribers.
  • filter:has_engagement — Filters out “ghost” tweets with zero interactions.

3. User & Account Filters

  • from:username — Sent by a specific user.
  • to:username — Sent as a reply to a specific user.
  • @username — Tweets mentioning a user.
  • list:12345 — Tweets from members of a specific List ID.
  • filter:news — Tweets linking to recognized news websites (e.g., NYT, BBC).

4. Content & Media Types

  • filter:images / filter:videos — Tweets with photos or videos.
  • filter:links — Tweets containing any URL.
  • card_name:animated_gif — Find tweets specifically with GIFs.
  • card_name:poll2choice_text_only — Find tweets containing polls.
  • filter:safe — Excludes NSFW or “Potentially Sensitive” content.

5. Date, Time & Precision

  • since:2025-01-01 / until:2025-01-31 — Date filtering (YYYY-MM-DD).
  • within_time:2d — Search within the last 2 days (supports h for hours, m for minutes).
  • since_id:ID / max_id:ID — Use Snowflake IDs for hyper-precise time boundaries.

6. Niche & Language Operators

  • lang:en — Filter by language (supports ru, es, fr, etc.).
  • lang:und — Undefined language (useful for finding tweets with only emojis or links).
  • #hashtag — Search for specific hashtags.
  • source:twitter_ads — Find tweets created via the X Ads platform.
  • geocode:40.71,-74.00,10km — Search tweets within a specific radius of coordinates (e.g., New York).

🧪 Practical Query Examples (Copy-Paste)

Example 1: Brand Analysis - Viral original content "Tesla" min_faves:500 filter:links lang:en -filter:replies Translates to: Original English tweets about “Tesla” with 500+ likes, containing a link. Example 2: OSINT - Finding news and media "breaking news" filter:images since:2026-02-01 filter:verified Translates to: Tweets containing “breaking news” with images from verified accounts since Feb 1st. Example 3: Customer Support - Finding user questions @AppleSupport "?" -from:AppleSupport Translates to: All tweets asking a question to Apple Support, excluding their own replies.

🛠 API Integration

In your Sorsa API request, simply pass the constructed string into the query field: JSON
{
  "query": "(artificial intelligence OR AI) min_faves:100 lang:en",
  "order": "latest",
  "next_cursor": ""
}

⏭ Next Steps