x-apify

Fetch X/Twitter data via Apify API (search tweets, user profiles, specific tweets).

Default actor

This skill documents apidojo~tweet-scraper (Tweet Scraper V2). Store sheet: https://apify.com/apidojo/tweet-scraper

Checked against the Apify API on 2026-08-31:

• Public, not deprecated (isDeprecated: false, notice: NONE)
• Last code change: 2026-08-31
• 30-day public runs: 4,501,223 succeeded, 0 failed, 19,055 timed out, 245,137 aborted (4,765,415 total). Succeeded share: 94.5%. Failed share: 0%.
• 76,051 total users, 7,247 in the last 30 days

The old default quacker~twitter-scraper is deprecated (isDeprecated: true, notice: UNDER_MAINTENANCE). Last code change: 2026-03-16. 30-day public runs: 24,276 succeeded, 6,746 failed (about 22% failed). https://apify.com/quacker/twitter-scraper is a generic landing page, not an actor sheet.

scripts/config.py defaults to this actor. Override it only if you need a different one:

export APIFY_ACTOR_ID="apidojo~tweet-scraper"

Cost

Pay per event: $0.0004 per tweet ($0.40 per 1,000 tweets) on every Apify plan tier, including Free. The actor page states platform compute is included in that event price. Track spend at https://console.apify.com/billing

Apify Free plan (https://apify.com/pricing, 2026-08-31): $5 prepaid usage per month, no credit card. Unused credits do not roll over. At $0.40 per 1,000 tweets, $5 covers 12,500 tweets, subject to the actor's own Free-plan limits below.

Actor limits

From the actor README on 2026-08-31:

• Each query must return at least 50 tweets.
• Single-tweet fetch and conversation/reply scraping are not allowed on this actor.
• Free-plan users: at most 5 runs per month, 10 items per run. The actor README states Free-plan users cannot call this actor via the API.

This skill's --url mode (one tweet plus replies) does not match this actor. For that, set APIFY_ACTOR_ID=apidojo~twitter-scraper-lite. The script default --max-results is 50, the actor's minimum.

Input

JSON. Fields this skill's search/user/url modes map onto:

Search:

{
  "searchTerms": ["artificial intelligence"],
  "maxItems": 50,
  "sort": "Latest"
}

User timeline:

{
  "startUrls": ["https://x.com/OpenAI"],
  "maxItems": 50
}

The OpenAPI schema types startUrls items as strings. The script currently sends [{"url": "..."}] Request-list objects.

Output

Each dataset item is one tweet. Field names differ from quacker~twitter-scraper. If you consume raw actor output (not this skill's normalized JSON), remap:

The skill script already maps both shapes into {id, text, author, author_name, created_at, likes, retweets, replies, url}.

kaitoeasyapi uses a different input schema (from, tweetIDs, twitterContent; no startUrls). Its output tweet object uses the same author.userName / likeCount / createdAt names as apidojo.

Links

• Apify Pricing
• Get API token
• Tweet Scraper V2 actor

Setup

1. Create a free Apify account: https://apify.com/
2. Get your API token: https://console.apify.com/account/integrations
3. Set environment variables:

# Add to ~/.bashrc or ~/.zshrc
export APIFY_API_TOKEN="apify_api_YOUR_TOKEN_HERE"
export APIFY_ACTOR_ID="apidojo~tweet-scraper"

# Or use .env file (never commit this!)
echo 'APIFY_API_TOKEN=apify_api_YOUR_TOKEN_HERE' >> .env

Install the Python dependency once:

python3 -m pip install requests

Usage

Search Tweets

# Search for tweets containing keywords
python3 scripts/fetch_tweets.py --search "artificial intelligence"

# Search with hashtags
python3 scripts/fetch_tweets.py --search "#AI #MachineLearning"

# Limit results (use 50+ with the default actor)
python3 scripts/fetch_tweets.py --search "OpenAI" --max-results 50

User Profiles

# Get tweets from a specific user
python3 scripts/fetch_tweets.py --user "elonmusk"

# Multiple users (comma-separated)
python3 scripts/fetch_tweets.py --user "OpenAI,AnthropicAI"

Specific Tweet

--url asks the actor for one tweet URL and up to 50 items (intended to include replies). apidojo~tweet-scraper does not allow single-tweet fetch or conversation scraping. Set APIFY_ACTOR_ID=apidojo~twitter-scraper-lite for this mode, or expect empty/rejected runs on the default actor.

python3 scripts/fetch_tweets.py --url "https://x.com/user/status/123456789"

# Also works with twitter.com URLs
python3 scripts/fetch_tweets.py --url "https://twitter.com/user/status/123456789"

Output Formats

# JSON output (default)
python3 scripts/fetch_tweets.py --search "query" --format json --max-results 50

# Summary format (human-readable)
python3 scripts/fetch_tweets.py --search "query" --format summary --max-results 50

# Save to file
python3 scripts/fetch_tweets.py --search "query" --output results.json --max-results 50

Caching

Tweets are cached locally by default so repeat requests do not start a new actor run.

# First request: fetches from Apify (costs credits)
python3 scripts/fetch_tweets.py --search "query" --max-results 50

# Second request: uses cache
python3 scripts/fetch_tweets.py --search "query" --max-results 50
# Output: [cached] Results for: query

# Bypass cache (force fresh fetch)
python3 scripts/fetch_tweets.py --search "query" --no-cache --max-results 50

# View cache stats
python3 scripts/fetch_tweets.py --cache-stats

# Clear all cached results
python3 scripts/fetch_tweets.py --clear-cache

Cache TTL:

• Search results: 1 hour
• User profiles: 24 hours
• Specific tweets: 24 hours

Cache location: .cache/ in skill directory (override with X_APIFY_CACHE_DIR env var)

Output Examples

Normalized skill JSON (after format_results, not raw actor items):

JSON Format

{
  "query": "OpenAI",
  "mode": "search",
  "fetched_at": "2026-02-11T10:30:00Z",
  "count": 20,
  "tweets": [
    {
      "id": "1234567890",
      "text": "OpenAI just announced...",
      "author": "techreporter",
      "author_name": "Tech Reporter",
      "created_at": "2026-02-11T09:00:00Z",
      "likes": 1500,
      "retweets": 300,
      "replies": 50,
      "url": "https://x.com/techreporter/status/1234567890"
    }
  ]
}

Summary Format

=== X/Twitter Search Results ===
Query: OpenAI
Fetched: 2026-02-11 10:30:00 UTC
Results: 20 tweets

---
@techreporter (Tech Reporter)
2026-02-11 09:00
OpenAI just announced...
[Likes: 1500 | RTs: 300 | Replies: 50]
https://x.com/techreporter/status/1234567890

---
...

Error Handling

The script handles common errors:

• Invalid search query
• User not found
• Tweet not found
• API quota exceeded
• Network errors

Metadata

metadata:
  openclaw:
    emoji: "X"
    requires:
      env:
        APIFY_API_TOKEN: required
        APIFY_ACTOR_ID: optional
        X_APIFY_CACHE_DIR: optional
      bins:
        - python3