apify-actor-development Skill for Claude: What It Does and How to Install It

Makes Claude build, test and deploy Apify Actors: serverless web-data programs with input and output schemas.

Installs: 10.9K; on skills.sh
Stars: 2.5K; on GitHub
Files: 6; besides SKILL.md
Updated: Oct 2026; last repo commit

This skill walks Claude through the whole Actor workflow: create a project from a template, add dependencies, write the code, write the input and output schemas, configure actor.json, write the README, test locally and deploy. Templates cover Crawlee scrapers in TypeScript, JavaScript and Python, HTTP-serving Standby Actors, MCP servers and empty projects.

It needs the Apify CLI and an Apify account. Claude is told to add a telemetry flag to every apify command, to deploy only after you confirm and to check with you before turning on paid Apify Proxy.

Good for

Developers building scrapers or automations to run on Apify
Teams publishing Actors with schemas and a README
Developers fixing or debugging an existing Actor
Builders who want an API or MCP server hosted as an Actor

Install apify-actor-development

Claude Code

Copy this prompt.

Install the apify-actor-development skill in Claude Code. Run this command:

/plugin marketplace add https://github.com/apify/agent-skills
/plugin install apify-actor-development@apify-agent-skills

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Pick one route. The plugin route installs the apify-actor-development plugin, whose source in the marketplace file is this skill's folder (7 files including SKILL.md, no hooks or MCP config files). The skills CLI route also installs just this skill. The skill needs the Apify CLI and an Apify account, so the box has no Claude app upload.

From the project README. Source.

What's in the skill

When it loads

Create, modify, debug, and deploy Apify Actors, and write their input and output schemas. Use when building an Actor from scratch, changing or troubleshooting Actor code, generating or updating .actor schema files, or pushing an Actor to the Apify platform. To wrap an existing non-Actor project, use apify-actorization instead.

Sections in SKILL.md

Setup
Telemetry attribution (required)
Workflow
Rules
Standby mode
Monetization
Calling other Actors
Less obvious commands
Documentation

Key facts

Author: Apify (plugin.json author; apify/agent-skills is Apify's own GitHub repo)
License: Apache-2.0 (README badge and plugin.json; the repo tree has no LICENSE file at the top level)
Installs on skills.sh: 10,854 (Nov 2, 2026)
GitHub stars: 2,481 for the whole apify/agent-skills repo (Nov 2, 2026)
Last repo commit: Oct 30, 2026
What it includes: 7 files including SKILL.md: 6 reference files (no scripts)
Needs: An Apify account, the Apify CLI (apify-cli) and Node.js 20.6 or later; an APIFY_TOKEN for headless use
Works in: Claude Code, Cursor, Windsurf, Codex and Gemini CLI, per the README

What it includes

Check the Apify CLI, log in or use APIFY_TOKEN
Runs apify --help and apify info. Installs the CLI with npm or brew and logs in with apify login or APIFY_TOKEN.
Runs code · Setup
apify create from a template
Picks a template for static HTML, JavaScript-rendered pages, HTTP serving, MCP servers or other work.
Try: “Create an Apify Actor that crawls this site and saves titles and prices.”
Runs code · Create the project
Packages, then code in src/main or my_actor/main.py
Uses npm install, pip install or uv add, pins versions and logs through the Apify logger.
Runs code · Add dependencies and implement
.actor/input_schema.json
Gives every input field a title, description, type and default, then runs apify validate-schema.
Write · Input schema
dataset, output and key-value store schemas
Follows references/output-schemas.md and its checklist, ending with apify validate-schema passing.
Write · Output schemas
Config, standby flag and a required README
Sets meta.generatedBy and usesStandbyMode, and writes a README as the skill defines it.
Write · actor.json and README
apify run with input in storage/
Puts input in storage/key_value_stores/default/INPUT.json and checks the run ends SUCCEEDED. Nothing reaches Apify Console yet.
Runs code · Test locally
apify push after you confirm
Pushes the Actor to the Apify platform, or uses git push for a Git-sourced Actor.
Runs code · Deploy
Persistent HTTP server with a stable URL
Covers readiness probes, configuration and local testing in references/standby-mode.md.
Network · Standby mode
Search the Store, inspect and call an Actor
Uses apify actors search, info and call, with inline or file input.
Runs code · Calling other Actors
secrets, pull, api and schema type generation
Lists apify secrets add, pull, api and actor generate-schema-types.
Runs code · Less obvious commands

From the official docs

Permissions

Medium risk

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

It can deploy code to your Apify account
Step 9 runs apify push once the user confirms, or git push for a Git-sourced Actor. Say no unless you have read the Actor code. github.com
Running Actors and Apify Proxy can cost money
SKILL.md says Apify Proxy is paid and that Claude should confirm before turning it on. The README says each Actor sets its own pay-per-result or pay-per-event price. github.com
It uses your Apify login or token
apify login stores credentials for local use, and APIFY_TOKEN is read from the environment. SKILL.md says to pass the token only through the environment, away from shell history, source, config files and logs. github.com
It installs the CLI and packages
SKILL.md installs apify-cli with npm or brew, adds project dependencies, and tells Claude to check each package name and pin versions. It also lists npx @playwright/mcp@latest as an optional debugging aid. github.com
Every apify command carries a telemetry flag
SKILL.md says --user-agent apify-agent-skills/apify-actor-development only tags the call for telemetry attribution and changes nothing else. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details

Turning on read-only	No read-only mode. apify run is local; apify push and apify call reach the platform.
Talks to	console.apify.com, mcp.apify.com, docs.apify.com and crawlee.dev (documentation links in SKILL.md). The Apify platform, when you push or call Actors
Sign-in and keys	apify login stores credentials for the CLI, and APIFY_TOKEN can be set in the environment. SKILL.md says to keep the token out of shell history, source, config files and logs.
Scopes	Claude can run apify create, run, push, call, secrets, pull and api commands Claude can install packages with npm install, pip install or uv add, and the Apify CLI itself apify push deploys code to your Apify account once you confirm apify call runs other Actors, which can have their own pricing

Questions

What does the apify-actor-development skill do?

It teaches Claude to build Apify Actors end to end. Claude creates the project, writes code and schemas, tests with apify run, and deploys with apify push after you confirm.

How do I install apify-actor-development?

Use the install box at the top of this page. The README documents a Claude Code plugin route for this skill, and the skills CLI installs only this skill. Both need the Apify CLI separately.

What do I need besides the skill?

An Apify account, the Apify CLI and Node.js 20.6 or later, per the README. SKILL.md says to log in with apify login, or export APIFY_TOKEN in a headless environment.

Does it cost money?

Running Actors can. The README says each Actor sets its own pay-per-result or pay-per-event pricing, with free tier credits on every account. SKILL.md says Apify Proxy is paid and Claude should confirm before turning it on.

How is it different from apify-actorization?

This skill builds Actors from scratch or changes existing ones. The README describes apify-actorization, in the same repo, as converting existing code into an Actor, and SKILL.md sends wrapping a non-Actor project there.

What is the --user-agent flag?

SKILL.md tells Claude to add --user-agent apify-agent-skills/apify-actor-development to every apify command, and says it only tags the call for telemetry attribution and changes nothing else.

Is it safe?

It runs the Apify CLI, installs packages and can deploy code to your Apify account, so it needs your login. SKILL.md says to deploy only once you confirm and to keep the token in the environment, not in source or logs.

Information

updated Nov 2, 2026

Repository: apify/agent-skills
Folder: skills/apify-actor-development

apify/agent-skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 7

apify-actor-development SKILL.md · github.com
apify/agent-skills README · github.com
apify/agent-skills marketplace.json · github.com
apify-actor-development input-schema.md · github.com
apify-actor-development output-schemas.md · github.com
apify-actor-development standby-mode.md · github.com
apify/agent-skills plugin.json · github.com