Search skills/

Apify Agent Skills Claude skills

Apify's official skills for scraping with its Actor marketplace and for building, converting and integrating Actors with the apify CLI.

apify/agent-skills 5 skills Browser automation & scrapingData & analytics
Our verdict Recommended with caveats

Official, well-referenced skills, but they only work with an Apify account whose Actors bill per result, and the repo has no detected license file despite the README's Apache-2.0 badge.

3/5 Depth expertise shipped Note Security Review before installing 10d Maintenance Updated 10 days ago 2.5K Adoption stars ? License no clear license

What Apify Agent Skills does

This is Apify's own skill pack for its web-scraping and automation platform. The headline skill, apify-ultimate-scraper, has the agent pick from about 130 curated Actors (ready-made scrapers for Instagram, TikTok, LinkedIn, Google Maps, Amazon, Airbnb and similar sites) using an index file and a set of workflow guides such as lead generation, competitor research and price monitoring. It then reads the Actor's input schema, runs it through the apify CLI with JSON output, and formats the results; if nothing in the index fits, it searches the full Apify Store.

The other skills are for developers. apify-actor-development creates, debugs and deploys Actors in JavaScript, TypeScript or Python and covers input and output schemas. apify-actorization wraps existing code or a CLI tool as an Actor. apify-integration-development guides building an official Apify integration for a workflow tool, agent plugin or framework package. A fifth skill, apify-generate-output-schema, is marked deprecated and points to the development skill. The five SKILL.md files average about 9 KB and the scraper alone has 16 reference files; there are no scripts or evals. Every apify command the skills issue carries a --user-agent apify-agent-skills/... flag, which the skills describe as telemetry attribution. You need the apify CLI and an Apify login or API token, and Actors bill per result or event.

Good for

  • Pulling data from social, maps, search and e-commerce sites through existing Apify Actors
  • Building and deploying a new Apify Actor with the official SDK patterns
  • Teams that already use Apify and want their agent to drive the CLI correctly

Not for

  • People who do not want a paid third-party scraping platform in the loop
  • Scraping where a site's terms or privacy law restrict collecting the data

What's inside depth 3/5

Five skills averaging 8.8 KB, with 30 reference and workflow files (16 in the scraper), but no scripts, tests or evals.

Skills
5
Reference docs
30
Scripts
0
Tests / evals
0
Templates
0
Avg SKILL.md
8.6 KB

Counted from the repository file tree. How we score depth.

Dependencies and costs

  • Apify account and the apify CLI (npm or Homebrew)
  • APIFY_TOKEN or apify login for authentication
  • Per-result or per-event Actor charges beyond the free credits
  • Docker for building and running Actors locally
  • Telemetry: The skills tell the agent to add --user-agent apify-agent-skills/<skill> to every apify command so Apify can attribute usage; they call it required and give no opt-out. The calls already go to Apify's own platform.

Security review

Review before installing

Official Apify skills for building Actors, converting projects into Actors, writing Apify integrations and running a universal scraper through the Apify CLI. Everything runs through the apify CLI against your Apify account, so runs can cost money; the skills require an attribution user-agent tag on every apify command, which only labels requests that already go to Apify. No allowed-tools or hooks are declared.

Worth knowing

  • Relies on a CLI or package we could not reviewNeeds the Apify CLI (npm install -g apify-cli or brew) which is not in the repo; the skills also mention npx @playwright/mcp and Docker for building Actors.skills/apify-ultimate-scraper/SKILL.md:16
  • Installs packages or toolsTells the agent to install apify-cli globally if missing (package manager only; apify-actorization/SKILL.md line 44 forbids curl-to-shell installs) and a Dockerfile example in references/cli-actorization.md bootstraps the ubi tool with curl.skills/apify-ultimate-scraper/SKILL.md:16
  • Needs an API key for its own serviceAuthenticates with APIFY_TOKEN (environment variable, apify login, or source .env) used only with Apify; the skills tell the agent to keep the token out of data pipelines and logs.skills/apify-ultimate-scraper/SKILL.md:26
  • Makes network requestsRuns paid Apify Actors that scrape platforms such as Instagram, TikTok, LinkedIn and Google Maps through the Apify platform (api.apify.com, mcp.apify.com); usage is billed per result or event beyond free credits.skills/apify-ultimate-scraper/SKILL.md:5

What we checked

The repository (45 files) on Oct 29, 2026. Not a full review: we read 2 of 3 code files; it relies on code outside the repository: Apify CLI (apify-cli npm package / Homebrew formula), Apify platform and the Actors it runs; npm package @playwright/mcp (mentioned, unpinned); Docker images used in Actor build examples.

NVIDIA SkillSpector static scan: 21 pattern hits (Agent Snooping 5, Data Exfiltration 5, MCP Rug Pull 4, Supply Chain 2, Tool Misuse 2, Anti-Refusal 1). The 21 hits are documentation: a security note that forbids curl | bash installs, a Dockerfile example that fetches the ubi bootstrap script, DELETE route lists for the Apify API, README skill links and unpinned npx mentions. The single script, generate_agents.py, only reads skill frontmatter and writes agents/AGENTS.md.

A review, not a certification. Skills change after we check them — read the files before installing.

License

README and badge say Apache-2.0, but GitHub reports no detected license file; check the repo before commercial reuse.

Can it run unattended?

Not really. The scraper skill runs one-off Actor jobs on request and the skills do not define budgets, stop conditions or resumable state, so unattended runs could rack up per-result charges.

Alternatives to Apify Agent Skills

Firecrawl CLI

firecrawl/cli

Firecrawl's command-line client and 13 agent skills for web search, scraping, crawling, interacting with pages and parsing files through the paid Firecrawl API.

With caveats Check first Depth 2/5

Firecrawl Skills

firecrawl/skills

Firecrawl's official catalog of 37 skills for web search, scraping, crawling and browser interaction, plus API integration and workflow recipes.

With caveats Check first Depth 3/5

agent-browser

vercel-labs/agent-browser

A native browser-automation CLI for agents, plus 11 skills that teach the snapshot-and-ref workflow for navigating, filling forms, scraping and testing.

Recommended Check first Depth 4/5

BrowserAct Skills

browser-act/skills

A browser-automation CLI skill with stealth browsers, CAPTCHA solving and handoff to a human, plus about 100 ready-made scraper skills for shops and lead sources.

With caveats Broad access Depth 3/5

FAQ

What are the Apify agent skills?

They are five skills from Apify for coding agents: a universal scraper that picks from 130+ curated Actors, plus skills for building Actors, wrapping existing code as Actors and building Apify integrations. A fifth, output-schema generation, is deprecated.

How do I install the Apify skills?

Run npx skills add https://github.com/apify/agent-skills --skill apify-ultimate-scraper (swap in another skill name as needed), or use the Claude Code plugin install shown in the repo. Then install the apify CLI with npm or Homebrew and run apify login.

Are the Apify skills free?

The skills are free to install, but they run Actors on the Apify platform. Actors use pay-per-result or pay-per-event pricing set by each Actor, and free-tier credits come with every account.

Is the Apify scraper skill safe to use?

We found no hidden network calls or credential theft in the files. The agent will use your Apify token to start billed runs and may scrape social platforms, so use a scoped token and check the target site's terms first.

Sources: GitHub repository (stars, license, file tree, last push Oct 19, 2026). Review based on the README and SKILL.md files as published.