Skills and Plugins
Power your agents with all the skills of Zyte API
Give Claude, Cursor and your own agents the same unblocking,
browser rendering and structured extraction that power Zyte API,
packaged as ready-to-install skills and plugins.
Skills and Plugins
Install Zyte in the agent you already use
Dedicated plugins for Claude Code, Codex and GitHub Copilot CLI, an extension for VS Code, and portable skills for everything else.
Claude Code
Claude Code Plugin
Turn prompts into production-ready web data pipelines. Skills to build a complete Scrapy project from just a URL and data type.
Codex CLI
Codex Plugin
Zyte's official plugin for Codex CLI. Describe what to extract in natural language and get a complete, tested Scrapy project.
GitHub Copilot CLI
GitHub CLI Plugin
Zyte skills for GitHub Copilot CLI. Go from a URL and data type to a complete, working Scrapy project without leaving your terminal.
VS Code extension
Web Scraping Copilot
VS Code extension that generates extraction logic and helps with project creation in your favorite IDE.
Any AI coding agent
Agent Skills
Portable skills that work with any AI coding agent. Add 15 years of web data expertise to whichever assistant your team already uses.
Claude Code
Codex CLI
Copilot CLI
VS Code
Any MCP clientWhy a plugin?
Claude tries to write a Scrapy spider With the Zyte skill, it gets it right
Bare Claude knows Python. It doesn't know Scrapy the way Zyte engineers do. The plugin teaches it async patterns, scrapy-poet page objects, the right libraries, real tests and anti-bot wired in. The spider that comes out of the prompt actually runs. Describe data. Get data.
| Your agent on its ownGeneric AI coding assistant | With Zyte skillsDomain-expert AI system | |
|---|---|---|
| A spider that runs first timeGet a runnable Scrapy spider for a new site, with no correction turns needed in our testing. | Gets modern Scrapy patterns wrongWrites simple, non-testable code and fails to include key add-ons. | Runs first time in our testingCorrect async Scrapy patterns, scrapy-poet page objects, targeted selectors, testable code with pytest. |
| Code that holds up to reviewExtraction shaped like a senior scraping engineer would write it. | Best-guess extractionOften decides your data shape for you, and hallucinates selectors, leaving you with questionable quality. | Built to Zyte's quality barCode shaped like the spiders Zyte engineers ship: accurate fields, clean patterns, scalable production spiders. |
| Anti-bot handled out of the boxBan handling wired in by default, with no unblocking layer to write yourself. | No anti-bot solutionYou wire one in by hand or watch the spider fail at the first ban. | Zyte API wired in by defaultBring your own if you prefer, but it works out of the box. |
| Testable code generationpytest to verify your code. | Doesn't know how to create HTML fixturesNo tests, or tests that don't run. | Tests scaffolded with the spiderpytest tests scaffolded against saved HTML fixtures. Tests fail? Prompt a fix and run again. |
| Scrapy stack awarenessThe right library at the right version, wired through correctly with our most important add-ons. | Produces generic ScrapyDoesn't know scrapy-poet, zyte-common-items, scrapy-zyte-api, or the add-ons system. | Reaches for the right library and versionWires them through the add-ons system, not legacy middleware. |
A spider that runs first time
Get a runnable Scrapy spider for a new site, with no correction turns needed in our testing.
Your agent on its ownGets modern Scrapy patterns wrongWrites simple, non-testable code and fails to include key add-ons.With Zyte skillsRuns first time in our testingCorrect async Scrapy patterns, scrapy-poet page objects, targeted selectors, testable code with pytest.Code that holds up to review
Extraction shaped like a senior scraping engineer would write it.
Your agent on its ownBest-guess extractionOften decides your data shape for you, and hallucinates selectors, leaving you with questionable quality.With Zyte skillsBuilt to Zyte's quality barCode shaped like the spiders Zyte engineers ship: accurate fields, clean patterns, scalable production spiders.Anti-bot handled out of the box
Ban handling wired in by default, with no unblocking layer to write yourself.
Your agent on its ownNo anti-bot solutionYou wire one in by hand or watch the spider fail at the first ban.With Zyte skillsZyte API wired in by defaultBring your own if you prefer, but it works out of the box.Testable code generation
pytest to verify your code.
Your agent on its ownDoesn't know how to create HTML fixturesNo tests, or tests that don't run.With Zyte skillsTests scaffolded with the spiderpytest tests scaffolded against saved HTML fixtures. Tests fail? Prompt a fix and run again.Scrapy stack awareness
The right library at the right version, wired through correctly with our most important add-ons.
Your agent on its ownProduces generic ScrapyDoesn't know scrapy-poet, zyte-common-items, scrapy-zyte-api, or the add-ons system.With Zyte skillsReaches for the right library and versionWires them through the add-ons system, not legacy middleware.
Tutorials
From installation to working spider in minutes
Claude Code Plugin
Introducing our Claude Code Plugin
See how the Zyte Claude Code plugin turns a website URL into a working, tested Scrapy scraper in under 15 minutes, ready to deploy and run.
Web Scraping Copilot
Web Scraping Copilot: VS Code extension for Scrapy projects
Web Scraping Copilot is built to reduce friction between idea and extraction. No context switching. No black boxes. No hidden automation.
FAQ
Skills and plugins FAQ
They teach your coding agent how to build production-ready web scrapers the way Zyte engineers do. Give your agent a URL and describe the data you want. It explores the site, proposes a schema for you to approve, then generates a complete Scrapy project with scrapy-poet page objects, pytest tests and Zyte API wired in. You don't need to write any scaffolding by hand.
Install the one for the agent you already use. The Claude Code, Codex and GitHub Copilot CLI plugins and the portable Agent Skills all run the same underlying skills and produce the same Scrapy project. The only differences are packaging and some agent-specific optimizations. Claude Code has the deepest integration. If you work in VS Code, Web Scraping Copilot brings extraction help into your IDE. If you want Zyte's fetching, extraction and Scrapy Cloud jobs available as tools inside any agent, connect Zyte MCP.
The skills and plugins are free. A Zyte API key isn't required, but we recommend one. Zyte API handles JavaScript rendering, anti-bot measures and reliability, so you'll get much better results on complex sites. Zyte API usage is billed separately. You'll also need the uv Python package manager, and the plugins will prompt you to install it if it's missing.
There are two separate stages. First, the agent fetches pages, discovers fields and iterates with you until you approve an extraction spec, which covers the schema and expected values. Second, it uses that approved spec to generate a complete Scrapy project with page objects, test fixtures and a spider. Code generation doesn't crawl the web. Crawling only happens when you run the spider locally or deploy it to Scrapy Cloud.
The extraction spec is the durable part of your project, and it isn't tied to any framework or agent. When a site changes, you re-run the spec stage or update the selectors in your page objects, and the rest of your project stays as it is. You can also use the spec on its own to feed plain Python, direct Zyte API extraction or a manual process.