Actor Router Examples and FAQ

A sales or research agent has LinkedIn profile URLs and needs names, titles, companies, summaries, and public profile data.

Recommended Actor

harvestapi/linkedin-profile-scraper

Agent full workflow

1. User request. User: Enrich these 30 LinkedIn profile URLs into a sales lead table.
2. Discover options. GET /v1/models?q=linkedin&limit=3 returns 143 matching Actors, including harvestapi/linkedin-profile-scraper apimaestro/linkedin-profile-detail dev_fusion/Linkedin-Profile-Scraper.
3. Inspect inputs. GET /v1/actors/harvestapi~linkedin-profile-scraper/schema shows example input, default memory/timeout, pricing hints, and normalized output schema.
4. Clarify missing constraints. If the user only provides company names, the agent asks for profile URLs, target region, required fields, and budget cap.
5. Validate without spending. POST /v1/runs with dry_run:true and body {"actor":"harvestapi/linkedin-profile-scraper","input":{"profileUrls":["https://www.linkedin.com/in/example/"],"maxItems":25},"max_total_charge_usd":1,"wait_for_finish":0,"dry_run":true}.
6. Start the real run. Remove dry_run. The gateway returns job_id, status_url, result_url, Actor metadata, and cost estimate.
7. Poll status. Call GET /v1/runs/{run_id} until status is SUCCEEDED, FAILED, ABORTED, or TIMED-OUT.
8. Read results. Call GET /v1/runs/{run_id}/result?limit=100 to receive normalized items plus raw upstream records.
9. Continue the conversation. Deliver a deduplicated lead table with missing fields, low-confidence profiles, and suggested companies for the next pass.

Run request

{
  "actor": "harvestapi/linkedin-profile-scraper",
  "input": {
    "profileUrls": [
      "https://www.linkedin.com/in/example/"
    ],
    "maxItems": 25
  },
  "max_total_charge_usd": 1,
  "wait_for_finish": 0
}

Returned result

{
  "object": "job_result",
  "schema": "social_profile.v1",
  "items": [
    {
      "url": "https://www.linkedin.com/in/example/",
      "title": "Example Person",
      "text": "Founder, AI tools",
      "raw": "{upstream actor item}"
    }
  ]
}

Advantage over direct Apify

Direct Apify calls require the agent to know the exact Actor, input fields, run flow, and dataset read path. Here it can search /v1/models?q=linkedin, run through /v1/runs, and receive normalized social_profile.v1 items.

Alternative Actors

harvestapi/linkedin-profile-scraper apimaestro/linkedin-profile-detail dev_fusion/Linkedin-Profile-Scraper

A content agent needs titles, text, transcript-like data, and metadata from YouTube videos or channels for summaries, competitor tracking, or RAG ingestion.

Recommended Actor

pintostudio/youtube-transcript-scraper

Agent full workflow

1. User request. User: Analyze these 5 YouTube videos, extract transcript points, and turn them into competitor insights.
2. Discover options. GET /v1/models?q=youtube&limit=3 returns 52 matching Actors, including streamers/youtube-scraper pintostudio/youtube-transcript-scraper streamers/youtube-channel-scraper.
3. Inspect inputs. GET /v1/actors/pintostudio~youtube-transcript-scraper/schema shows example input, default memory/timeout, pricing hints, and normalized output schema.
4. Clarify missing constraints. The agent asks whether the input is video URLs or channel URLs, desired output language, transcript retention, and max items per video.
5. Validate without spending. POST /v1/runs with dry_run:true and body {"actor":"pintostudio/youtube-transcript-scraper","input":{"urls":["https://www.youtube.com/watch?v=dQw4w9WgXcQ"],"maxItems":10},"max_total_charge_usd":1,"wait_for_finish":0,"dry_run":true}.
6. Start the real run. Remove dry_run. The gateway returns job_id, status_url, result_url, Actor metadata, and cost estimate.
7. Poll status. Call GET /v1/runs/{run_id} until status is SUCCEEDED, FAILED, ABORTED, or TIMED-OUT.
8. Read results. Call GET /v1/runs/{run_id}/result?limit=100 to receive normalized items plus raw upstream records.
9. Continue the conversation. Deliver summaries, topic clusters, citeable snippets, and normalized text/url/title records for ingestion.

Run request

{
  "actor": "pintostudio/youtube-transcript-scraper",
  "input": {
    "urls": [
      "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
    ],
    "maxItems": 10
  },
  "max_total_charge_usd": 1,
  "wait_for_finish": 0
}

Returned result

{
  "object": "job_result",
  "schema": "social_profile.v1",
  "items": [
    {
      "url": "https://www.youtube.com/watch?v=...",
      "title": "Video title",
      "text": "Transcript text...",
      "raw": "{upstream actor item}"
    }
  ]
}

Advantage over direct Apify

YouTube becomes a capability entry point instead of a hard-coded Actor dependency. The agent can inspect available Actors, quality signals, and cost controls before running.

Alternative Actors

streamers/youtube-scraper pintostudio/youtube-transcript-scraper streamers/youtube-channel-scraper

A market or investment agent needs public posts by keyword, account, or topic for trend, sentiment, or event tracking.

Recommended Actor

apidojo/tweet-scraper

Agent full workflow

1. User request. User: Monitor recent X/Twitter discussion about AI agents and find high-engagement opinions.
2. Discover options. GET /v1/models?q=twitter&limit=3 returns 41 matching Actors, including apidojo/tweet-scraper kaitoeasyapi/twitter-x-data-tweet-scraper-pay-per-result-cheapest xquik/x-tweet-scraper.
3. Inspect inputs. GET /v1/actors/apidojo~tweet-scraper/schema shows example input, default memory/timeout, pricing hints, and normalized output schema.
4. Clarify missing constraints. The agent asks for keywords, time range, language, exclusions, and whether to focus on verified or high-follower accounts.
5. Validate without spending. POST /v1/runs with dry_run:true and body {"actor":"apidojo/tweet-scraper","input":{"searchTerms":["AI agent"],"maxItems":100},"max_total_charge_usd":2,"wait_for_finish":0,"dry_run":true}.
6. Start the real run. Remove dry_run. The gateway returns job_id, status_url, result_url, Actor metadata, and cost estimate.
7. Poll status. Call GET /v1/runs/{run_id} until status is SUCCEEDED, FAILED, ABORTED, or TIMED-OUT.
8. Read results. Call GET /v1/runs/{run_id}/result?limit=100 to receive normalized items plus raw upstream records.
9. Continue the conversation. Deliver high-engagement posts, opinion clusters, trend judgment, and accounts worth tracking.

Run request

{
  "actor": "apidojo/tweet-scraper",
  "input": {
    "searchTerms": [
      "AI agent"
    ],
    "maxItems": 100
  },
  "max_total_charge_usd": 2,
  "wait_for_finish": 0
}

Returned result

{
  "object": "job_result",
  "schema": "social_profile.v1",
  "items": [
    {
      "url": "https://x.com/user/status/...",
      "title": "@user",
      "text": "Post text...",
      "raw": "{upstream actor item}"
    }
  ]
}

Advantage over direct Apify

Results are normalized into stable url/title/text/raw fields, so the agent does not need a custom parser for every Twitter/X Actor.

Alternative Actors

apidojo/tweet-scraper kaitoeasyapi/twitter-x-data-tweet-scraper-pay-per-result-cheapest xquik/x-tweet-scraper

A sales agent wants local businesses in a city and category, including address, website, phone, ratings, review count, and category.

Recommended Actor

compass/crawler-google-places

Agent full workflow

1. User request. User: Find coffee shops near Austin for outbound sales, preferring websites and high ratings.
2. Discover options. GET /v1/models?q=google&limit=3 returns 71 matching Actors, including apify/google-search-scraper compass/crawler-google-places compass/Google-Maps-Reviews-Scraper.
3. Inspect inputs. GET /v1/actors/compass~crawler-google-places/schema shows example input, default memory/timeout, pricing hints, and normalized output schema.
4. Clarify missing constraints. The agent asks for radius, category keywords, max businesses, and whether phone, website, or rating thresholds are required.
5. Validate without spending. POST /v1/runs with dry_run:true and body {"actor":"compass/crawler-google-places","input":{"searchString":"coffee near Austin","maxCrawledPlaces":5,"proxyConfig":{"useApifyProxy":true}},"max_total_charge_usd":1,"wait_for_finish":0,"dry_run":true}.
6. Start the real run. Remove dry_run. The gateway returns job_id, status_url, result_url, Actor metadata, and cost estimate.
7. Poll status. Call GET /v1/runs/{run_id} until status is SUCCEEDED, FAILED, ABORTED, or TIMED-OUT.
8. Read results. Call GET /v1/runs/{run_id}/result?limit=100 to receive normalized items plus raw upstream records.
9. Continue the conversation. Deliver a business_lead.v1 table ranked by rating, review count, and website availability, plus next search terms.

Run request

{
  "actor": "compass/crawler-google-places",
  "input": {
    "searchString": "coffee near Austin",
    "maxCrawledPlaces": 5,
    "proxyConfig": {
      "useApifyProxy": true
    }
  },
  "max_total_charge_usd": 1,
  "wait_for_finish": 0
}

Returned result

{
  "object": "job_result",
  "schema": "business_lead.v1",
  "items": [
    {
      "title": "Trippy Buck Coffee",
      "website": "https://linktr.ee/TrippyBuckCoffee",
      "address": "720 Brazos St, Austin, TX 78701",
      "reviews_count": 36
    }
  ]
}

Advantage over direct Apify

This path has been verified live: /v1/runs produced a SUCCEEDED job and /v1/runs/{run_id}/result returned business_lead.v1 items.

Alternative Actors

apify/google-search-scraper compass/crawler-google-places compass/Google-Maps-Reviews-Scraper

A brand agent needs public posts, bios, engagement hints, and content links by account or hashtag.

Recommended Actor

apify/instagram-scraper

Agent full workflow

1. User request. User: Collect recent OpenAI Instagram content and summarize themes and engagement direction.
2. Discover options. GET /v1/models?q=instagram&limit=3 returns 95 matching Actors, including apify/instagram-scraper apify/instagram-profile-scraper apify/instagram-post-scraper.
3. Inspect inputs. GET /v1/actors/apify~instagram-scraper/schema shows example input, default memory/timeout, pricing hints, and normalized output schema.
4. Clarify missing constraints. The agent asks for accounts or hashtags, time range, post count, and whether image links and full captions are needed.
5. Validate without spending. POST /v1/runs with dry_run:true and body {"actor":"apify/instagram-scraper","input":{"directUrls":["https://www.instagram.com/openai/"],"resultsLimit":30},"max_total_charge_usd":2,"wait_for_finish":0,"dry_run":true}.
6. Start the real run. Remove dry_run. The gateway returns job_id, status_url, result_url, Actor metadata, and cost estimate.
7. Poll status. Call GET /v1/runs/{run_id} until status is SUCCEEDED, FAILED, ABORTED, or TIMED-OUT.
8. Read results. Call GET /v1/runs/{run_id}/result?limit=100 to receive normalized items plus raw upstream records.
9. Continue the conversation. Deliver post records, themes, engagement signals, and a reusable content strategy summary.

Run request

{
  "actor": "apify/instagram-scraper",
  "input": {
    "directUrls": [
      "https://www.instagram.com/openai/"
    ],
    "resultsLimit": 30
  },
  "max_total_charge_usd": 2,
  "wait_for_finish": 0
}

Returned result

{
  "object": "job_result",
  "schema": "social_profile.v1",
  "items": [
    {
      "url": "https://www.instagram.com/p/...",
      "title": "openai",
      "text": "Caption text...",
      "raw": "{upstream actor item}"
    }
  ]
}

Advantage over direct Apify

The directory exposes routing signals like users, runs, revenue, failure rate, and agentic support, so the agent can choose a higher-quality Actor before running.

Alternative Actors

apify/instagram-scraper apify/instagram-profile-scraper apify/instagram-post-scraper

A growth agent needs creator profiles, videos, captions, play counts, or engagement signals from TikTok.

Recommended Actor

clockworks/tiktok-scraper

Agent full workflow

1. User request. User: Find popular TikTok videos from a creator category and summarize what makes them work.
2. Discover options. GET /v1/models?q=tiktok&limit=3 returns 77 matching Actors, including clockworks/tiktok-scraper clockworks/free-tiktok-scraper clockworks/tiktok-profile-scraper.
3. Inspect inputs. GET /v1/actors/clockworks~tiktok-scraper/schema shows example input, default memory/timeout, pricing hints, and normalized output schema.
4. Clarify missing constraints. The agent asks for creator list or keywords, region/language, video count, and whether play/like metrics are required.
5. Validate without spending. POST /v1/runs with dry_run:true and body {"actor":"clockworks/tiktok-scraper","input":{"profiles":["tiktok"],"resultsPerPage":20},"max_total_charge_usd":2,"wait_for_finish":0,"dry_run":true}.
6. Start the real run. Remove dry_run. The gateway returns job_id, status_url, result_url, Actor metadata, and cost estimate.
7. Poll status. Call GET /v1/runs/{run_id} until status is SUCCEEDED, FAILED, ABORTED, or TIMED-OUT.
8. Read results. Call GET /v1/runs/{run_id}/result?limit=100 to receive normalized items plus raw upstream records.
9. Continue the conversation. Deliver a video table, caption themes, engagement ranking, and reusable script or topic patterns.

Run request

{
  "actor": "clockworks/tiktok-scraper",
  "input": {
    "profiles": [
      "tiktok"
    ],
    "resultsPerPage": 20
  },
  "max_total_charge_usd": 2,
  "wait_for_finish": 0
}

Returned result

{
  "object": "job_result",
  "schema": "social_profile.v1",
  "items": [
    {
      "url": "https://www.tiktok.com/@...",
      "title": "Creator",
      "text": "Video caption...",
      "raw": "{upstream actor item}"
    }
  ]
}

Advantage over direct Apify

The same /v1/runs protocol covers Instagram, TikTok, YouTube, and Twitter/X, so agents do not need four separate Apify workflows.

Alternative Actors

clockworks/tiktok-scraper clockworks/free-tiktok-scraper clockworks/tiktok-profile-scraper

An ecommerce agent needs prices, availability, images, ratings, and competing listings from search terms or product URLs.

Recommended Actor

junglee/Amazon-crawler

Agent full workflow

1. User request. User: Monitor Amazon competitor prices and availability for wireless headphones.
2. Discover options. GET /v1/models?q=amazon&limit=3 returns 26 matching Actors, including junglee/Amazon-crawler axesso_data/amazon-reviews-scraper junglee/free-amazon-product-scraper.
3. Inspect inputs. GET /v1/actors/junglee~Amazon-crawler/schema shows example input, default memory/timeout, pricing hints, and normalized output schema.
4. Clarify missing constraints. The agent asks for marketplace country, keywords or ASINs, price range, max products, and whether recurring comparison is needed.
5. Validate without spending. POST /v1/runs with dry_run:true and body {"actor":"junglee/Amazon-crawler","input":{"search":"wireless headphones","maxItems":50},"max_total_charge_usd":2,"wait_for_finish":0,"dry_run":true}.
6. Start the real run. Remove dry_run. The gateway returns job_id, status_url, result_url, Actor metadata, and cost estimate.
7. Poll status. Call GET /v1/runs/{run_id} until status is SUCCEEDED, FAILED, ABORTED, or TIMED-OUT.
8. Read results. Call GET /v1/runs/{run_id}/result?limit=100 to receive normalized items plus raw upstream records.
9. Continue the conversation. Deliver product_price.v1 results as a competitor table sorted by price, rating, availability, and image links.

Run request

{
  "actor": "junglee/Amazon-crawler",
  "input": {
    "search": "wireless headphones",
    "maxItems": 50
  },
  "max_total_charge_usd": 2,
  "wait_for_finish": 0
}

Returned result

{
  "object": "job_result",
  "schema": "product_price.v1",
  "items": [
    {
      "title": "Wireless headphones",
      "price": "$59.99",
      "availability": "In stock",
      "image": "https://..."
    }
  ]
}

Advantage over direct Apify

Results enter product_price.v1, which makes price comparison, alerting, and spreadsheet export simpler than handling each Actor schema separately.

Alternative Actors

junglee/Amazon-crawler axesso_data/amazon-reviews-scraper junglee/free-amazon-product-scraper

A recruiting or research agent wants public jobs, companies, locations, salary hints, and posting dates.

Recommended Actor

apimaestro/linkedin-profile-batch-scraper-no-cookies-required

Agent full workflow

1. User request. User: Find remote AI engineer jobs and summarize company and salary signals.
2. Discover options. GET /v1/models?q=jobs&limit=3 returns 111 matching Actors, including apimaestro/linkedin-profile-batch-scraper-no-cookies-required curious_coder/linkedin-jobs-scraper misceres/indeed-scraper.
3. Inspect inputs. GET /v1/actors/apimaestro~linkedin-profile-batch-scraper-no-cookies-required/schema shows example input, default memory/timeout, pricing hints, and normalized output schema.
4. Clarify missing constraints. The agent asks for region, role keywords, seniority, whether salary is required, and max postings.
5. Validate without spending. POST /v1/runs with dry_run:true and body {"capability":"job_posting_extraction","input":{"query":"AI engineer remote","maxItems":100},"max_total_charge_usd":2,"wait_for_finish":0,"dry_run":true}.
6. Start the real run. Remove dry_run. The gateway returns job_id, status_url, result_url, Actor metadata, and cost estimate.
7. Poll status. Call GET /v1/runs/{run_id} until status is SUCCEEDED, FAILED, ABORTED, or TIMED-OUT.
8. Read results. Call GET /v1/runs/{run_id}/result?limit=100 to receive normalized items plus raw upstream records.
9. Continue the conversation. Deliver a job_posting.v1 table sorted by company, location, salary, posting date, and apply URL.

Run request

{
  "capability": "job_posting_extraction",
  "input": {
    "query": "AI engineer remote",
    "maxItems": 100
  },
  "max_total_charge_usd": 2,
  "wait_for_finish": 0
}

Returned result

{
  "object": "job_result",
  "schema": "job_posting.v1",
  "items": [
    {
      "title": "AI Engineer",
      "company": "Example Inc.",
      "location": "Remote",
      "salary": "$150k-$220k"
    }
  ]
}

Advantage over direct Apify

The agent can call by capability and let the router pick a matching Actor. Direct Apify use usually requires knowing the exact Actor first.

Alternative Actors

apimaestro/linkedin-profile-batch-scraper-no-cookies-required curious_coder/linkedin-jobs-scraper misceres/indeed-scraper

A support or brand agent needs public reviews from places like Google Maps, Tripadvisor, or Trustpilot to summarize themes and root causes.

Recommended Actor

compass/Google-Maps-Reviews-Scraper

Agent full workflow

1. User request. User: Scrape public reviews for this business and tell me what negative reviews focus on.
2. Discover options. GET /v1/models?q=reviews&limit=3 returns 37 matching Actors, including compass/Google-Maps-Reviews-Scraper maxcopell/tripadvisor-reviews apify/facebook-reviews-scraper.
3. Inspect inputs. GET /v1/actors/compass~Google-Maps-Reviews-Scraper/schema shows example input, default memory/timeout, pricing hints, and normalized output schema.
4. Clarify missing constraints. The agent asks for business URL, platform, review count, whether to focus on low ratings, and desired dimensions.
5. Validate without spending. POST /v1/runs with dry_run:true and body {"capability":"review_scraping","input":{"url":"https://www.google.com/maps/place/example","maxItems":100},"max_total_charge_usd":2,"wait_for_finish":0,"dry_run":true}.
6. Start the real run. Remove dry_run. The gateway returns job_id, status_url, result_url, Actor metadata, and cost estimate.
7. Poll status. Call GET /v1/runs/{run_id} until status is SUCCEEDED, FAILED, ABORTED, or TIMED-OUT.
8. Read results. Call GET /v1/runs/{run_id}/result?limit=100 to receive normalized items plus raw upstream records.
9. Continue the conversation. Deliver review.v1 items, negative themes, representative quotes, and improvement priorities.

Run request

{
  "capability": "review_scraping",
  "input": {
    "url": "https://www.google.com/maps/place/example",
    "maxItems": 100
  },
  "max_total_charge_usd": 2,
  "wait_for_finish": 0
}

Returned result

{
  "object": "job_result",
  "schema": "review.v1",
  "items": [
    {
      "rating": 5,
      "author": "Reviewer",
      "date": "2026-07-03",
      "text": "Great service..."
    }
  ]
}

Advantage over direct Apify

The agent can route through review_scraping and retrieve review.v1, avoiding per-platform polling and parsing code.

Alternative Actors

compass/Google-Maps-Reviews-Scraper maxcopell/tripadvisor-reviews apify/facebook-reviews-scraper

A B2B agent has a company name or domain and needs public website, industry, employee, contact, or LinkedIn company data.

Recommended Actor

harvestapi/linkedin-company

Agent full workflow

1. User request. User: Enrich these company names with websites, LinkedIn company pages, and contact signals.
2. Discover options. GET /v1/models?q=company&limit=3 returns 611 matching Actors, including apify/google-search-scraper apify/instagram-profile-scraper compass/crawler-google-places.
3. Inspect inputs. GET /v1/actors/harvestapi~linkedin-company/schema shows example input, default memory/timeout, pricing hints, and normalized output schema.
4. Clarify missing constraints. The agent asks for company list, target country, whether emails, employee count, or industry are needed, and budget cap.
5. Validate without spending. POST /v1/runs with dry_run:true and body {"actor":"harvestapi/linkedin-company","input":{"companies":["OpenAI"],"maxItems":25},"max_total_charge_usd":2,"wait_for_finish":0,"dry_run":true}.
6. Start the real run. Remove dry_run. The gateway returns job_id, status_url, result_url, Actor metadata, and cost estimate.
7. Poll status. Call GET /v1/runs/{run_id} until status is SUCCEEDED, FAILED, ABORTED, or TIMED-OUT.
8. Read results. Call GET /v1/runs/{run_id}/result?limit=100 to receive normalized items plus raw upstream records.
9. Continue the conversation. Deliver a company_profile.v1 table with raw data for review and companies that deserve a deeper second pass.

Run request

{
  "actor": "harvestapi/linkedin-company",
  "input": {
    "companies": [
      "OpenAI"
    ],
    "maxItems": 25
  },
  "max_total_charge_usd": 2,
  "wait_for_finish": 0
}

Returned result

{
  "object": "job_result",
  "schema": "company_profile.v1",
  "items": [
    {
      "company": "OpenAI",
      "domain": "openai.com",
      "contacts": [
        "public contact item"
      ],
      "raw": "{upstream actor item}"
    }
  ]
}

Advantage over direct Apify

The gateway keeps raw upstream data while exposing company_profile.v1 summary fields, giving the agent both stable fields and full fallback detail.

Alternative Actors

apify/google-search-scraper apify/instagram-profile-scraper compass/crawler-google-places