The field index glossary.

The terms that decide whether a data tool fits: what an Actor is, what a schema promises, how the price is charged and where access ends. Each term links to the page that applies it.

Use the definitions carefully

A field name is not a guarantee of a populated value. A published success rate is not a correctness score. A public page is not permission to collect or reuse everything on it. These definitions keep those distinctions visible.

Catalogue data checked 2026-11-12. Read the buying guides or browse all Actors.

Buying the data

Actor

A packaged program on Apify that accepts structured input, runs on the platform and writes output to a dataset. A Store listing is the public page for one Actor. Open the related page.

Store listing

The public distribution page for an Actor. It carries the description, pricing information, input and output schemas, build date and public usage statistics. Open the related page.

Catalogue

The set of Apify Store Actors this site maps and recommends. It is a curated selection and the source for the recommendation box, field finder and cost estimator. Open the related page.

Input schema

The published description of the fields an Actor accepts, including required fields, types and defaults. A run that does not match the schema can fail before it returns data. Open the related page.

Output schema

The published description of the fields an Actor writes to its dataset. It shows names and types, but not whether every record has a non-empty value. Open the related page.

Dataset

The structured output of a run, usually available as JSON, CSV or Excel. A dataset can contain nested objects and arrays, not only flat columns. Open the related page.

Field coverage

The share or count of the output fields a tool publishes for a target use case. On this site it is measured from the published schema for the target. Open the related page.

Price per 1,000

The published primary-result rate for 1,000 records or events. It excludes platform usage, start events, add-ons and plan differences. Open the related page.

Pay per event

A pricing model where a run charges for a defined event, such as one result, one page or one actor start. The bill depends on which events occur. Open the related page.

Success rate

The share of runs that completed over a window, as reported by the platform. It is not a measure of data correctness or field completeness. Open the related page.

Freshness

How recently the tool was built or updated. It is a useful signal that maintenance is active, not proof that the current site still works. Open the related page.

Access and limits

Official API

A documented interface published by the source site or platform. It has its own authentication, pricing, rate limits and terms. Open the related page.

Partner API

An official interface available only through an approved partner, research or commercial programme. Access is not self-serve. Open the related page.

Authwall

A login redirect that blocks an unauthenticated request. It is an access restriction, not a parser bug. Do not bypass it. Open the related page.

Rate limit

A cap on requests over a time window. An HTTP 429 or a platform-specific throttle code usually means the caller should back off rather than retry immediately. Open the related page.

Robots.txt

A file that tells crawlers which paths the operator prefers not to have crawled. It is one access signal, not a complete legal analysis. Open the related page.

Pagination

The mechanism that splits a result set into pages. A result ceiling can stop a list before it is complete, even when pagination exists. Open the related page.

Proxy

An intermediary that sends a request from a different network address. Proxies can change access and cost, but they do not create permission to collect restricted data. Open the related page.

Agents and automation

MCP

Model Context Protocol, a standard interface that lets an AI client call external tools and data sources. An Apify MCP server can expose one Actor or a wider tool set. Open the related page.

Agent skill

A packaged instruction file, usually with a script and example input, that tells an agent when and how to perform a bounded task. Open the related page.

n8n workflow

A node-based automation that connects a data source, transformations and a destination. The artifacts on Recipe for Agents are validated for structure, not executed. Open the related page.

Run-sync

An API pattern that starts an Actor run and waits for the result in the same request. It is convenient for small jobs and less suitable for long-running runs. Open the related page.

All terms

Actor, Store listing, Catalogue, Input schema, Output schema, Dataset, Field coverage, Price per 1,000, Pay per event, Success rate, Freshness, Official API, Partner API, Authwall, Rate limit, Robots.txt, Pagination, Proxy, MCP, Agent skill, n8n workflow, Run-sync.

What is an Actor?How to read a schemaHow pricing worksHow access works

Glossary updated 2026-11-10Catalogue checked 2026-11-12