Apify Actors Checklist
Things to ensure before publishing an actor.
Identity
Section titled “Identity”All identity fields should be filled in. Technical names are assumed to be permanent because they are used for URLs.
Identity fields:
- Name: public name
- Description: contains actor’s key details and function using verbs, around 300 characters.
- Technical Name: permanent identifier on Apify in kebab case
- SEO Name: short name with keywords for discovery
- SEO Description: short description with keywords for promotion, around 150 characters
Documentation
Section titled “Documentation”Comprehensive README + input/output schemas and examples.
Actor README
Section titled “Actor README”The Actor README functions like a landing page plus documentation overview. It should be written in a way intended for the target user base of the actor. It should be structured in markdown and basic HTML (for tables, etc) such that it produces a clean Table of Contents on the platform page.
A README should:
- contain relevant language for SEO
- make a good first impression
- provide extended instructions for users
- act as a support hub using Apify issues
A README should contain:
- 2-3 paragraphs description that explains what the actor is and what it can do
- input and output shape
- costs structure to set expectations
- cross-promote and integration with other actors
- FAQ
Pricing
Section titled “Pricing”Pay-per-event Model (PPE)
Section titled “Pay-per-event Model (PPE)”https://docs.apify.com/platform/actors/publishing/monetize/pay-per-event
It’s highly recommended to use a Pay-per-event (PPE) model, especially for actors that orchestrate external APIs. With this model, it’s crucial to be transparent on how the actor works and what it’s doing, as well as fairly account for allocated resources. Quality and transparent telemetry/logging is key.
Concurrency and Resource Allocation
Section titled “Concurrency and Resource Allocation”https://docs.apify.com/platform/actors/running/usage-and-resources
Actor runs are allocated memory (RAM), compute (CPU) and storage at specific ratios. CPU and storage are allocated based on the memory set for the run.
It’s common practice for actors to take multiple inputs e.g. a list of URLs, as well as use concurrency internally in processing each input. It is highly recommended to manage the actor’s concurrency based on available resources (Apify exposes this information via environment variables) and the resource consumption behavior of the actor. Benchmarking, sensible allocation defaults and internal concurrency management are key.