Choose an Apify Actor by Input and Result

Choose an Apify Actor by the data you already have and the result you need. A Twitter account whose followers you want is a different input from a list of usernames whose emails you need. A Zillow ZIP search is different from a lookup of known property IDs. Pick that task first, then open its dedicated walkthrough.

This directory covers 26 x_guru Actors and links to the individual Dievio guides. Keep each source separate. Only connect Actors when the first result provides a meaningful input for the next step, such as follower usernames for email lookup or ZPIDs for property details.

1. Write down what you have: search terms, profile URLs, usernames, business domains, property IDs or a watchlist.
2. Define the result: one business, contact match, follower connection, review, property, creative or change event.
3. Choose discovery, enrichment or monitoring: discovery builds a list; enrichment adds fields to known identifiers; monitoring compares repeat observations.
4. Open the matching guide: use its current input fields, output mapping and continuation rules.
5. Run a bounded sample: review source coverage and actual rows before expanding.

Similar names do not guarantee interchangeable schemas. Check Input, Output and Pricing for the exact Actor you choose. Do not carry a JSON field from another tool simply because both collect data from the same website.

This example is the Google Maps Scraper's read-only field reference. It illustrates why search input, location and result limits must be checked against the chosen Actor. Your editable form opens in Apify Console after Try for free.

Use business discovery when you have a category and area. Use email extraction when your goal is business contact matches, and review collection coverage separately from email coverage. Use review extraction when you already know the places and need individual review rows.

Enrich an existing Maps dataset with website emails, retaining Place ID and source attribution.

These tools start with known public business websites. They crawl for available contact information; they do not discover every business in a city. Keep source URLs with found fields, and retain websites where nothing was found so your report shows the original denominator.

Profile lookup starts with known public LinkedIn profile URLs. B2B search starts with company and role filters and discovers matching contacts. An independent search is different from exporting a saved Apollo list or using Apollo's authorized API.

For a combined result, join LinkedIn email and phone lookups by profile URL. For the separate product decision, compare Apollo export, Apollo API and independent B2B search.

Choose the followers Actor to discover an account's audience. Choose the email Actor when you already have the usernames to check. Use the two together only for the specific audience-to-email workflow. Follower collection does not itself find emails.

The two-Actor follower-to-email walkthrough preserves audience membership while enriching unique handles. The username email guide explains both the direct-lookup and audience-first routes.

Use search tools for geographic discovery, details for known property identifiers, and the rental or sold tools for those specific inventories. Regional ZHVI/ZORI series are monthly indices, not property-level estimates or a substitute for listing data.

If you already completed discovery, pass a Zillow Search dataset to Property Details. Join by ZPID and keep every original search row and its geographic context.

Collect public creatives for a confirmed advertiser or reviewed search. These sources do not provide private campaign performance. Each monitor has its own filters, coverage rules and completed baseline.

Start with a specific product variant and a verified market. A whole-store catalog is a broader scope. Availability is not a guaranteed inventory quantity, and change records are not automatically delivered notifications.

Monitor supported public employer boards for job-posting activity. This collects vacancies and their changes, rather than applicants or employee contact details. A removed posting does not establish that someone was hired.

Deduplicate lookup inputs when appropriate, but keep the original relationship table. One person can belong to several audiences and one property can appear in several searches. Join enrichment back without dropping unmatched rows or inventing a second-stage success.

Do not combine unrelated actors into one recipe merely because they share an Apify account. Twitter audiences, Zillow properties, ad creatives, retail offers and published jobs answer different user questions and use different identifiers.

Your compatible AI agent can inspect Actors, start bounded runs and retrieve results through Apify MCP. Connect https://mcp.apify.com, authorize your account and follow the official setup instructions. Name the actual Actor rather than asking the agent to guess from a broad topic.

Use Apify MCP to inspect these Actors:
1. x_guru/twitter-follower-scraper
2. x_guru/twitter-email-scraper
Plan a Twitter audience-to-email workflow
for my supplied public target accounts.
Check current schemas, input minimums,
output identifiers and live pricing.
Show the chosen limits and the two budgets.
Explain the handle join and unmatched rows.
Before any approved execution, set
callOptions.maxTotalChargeUsd per run.
Do not start paid runs until I approve
this scope. Do not substitute unrelated
Actors or promise an email for every user.

This prompt deliberately plans before execution because email lookup has its own batch requirements and costs. Each dedicated guide supplies a more specific prompt for its source. MCP access also depends on Actor permissions and your account's access; full-permission Actors are excluded by Apify MCP.

The agent should use fetch-actor-details before execution, call-actor to start, get-actor-run for unfinished work and get-dataset-items with pagination for results. Source reports use get-key-value-store-record. A summary of a few sample rows is not a complete exported dataset.

• Input limit: may be per target, per run, per collection or a check-unit ceiling.
• Spending cap: belongs in run options and can stop before the requested output count.
• Source completion: check reports and failures, including targets not visited.
• Continuation: use the selected Actor's documented next input; field names differ.
• Result coverage: distinguish searched records, saved observations and available contact matches.
• Repeat checks: retain stable monitor identity and scope, and inspect source health before interpreting disappearance.

Large allowed limits are not throughput or completeness guarantees. Free account tiers can still consume available credits, and continuation usually does not reset collection-wide allowances. Use each Actor's live Pricing tab rather than applying one tool's rate to another.

For a direct application integration or an existing saved dataset, follow how to run an Actor and export its dataset. It covers Console, API, MCP, run identity, CSV export and paginated JSON without restarting a completed collection.