Scrapegraph-ai
AI Agents & AssistantsA Python library that builds web‑scraping pipelines using LLMs and graph logic, letting you extract data from websites or local documents with just a prompt.
Features
- LLM‑driven scraping graphs to auto‑generate extraction pipelines
- Supports multiple file formats (HTML, XML, JSON, Markdown, etc.) and both web URLs and local files
- Provides Python SDK with configurable LLM settings (e.g., Ollama, OpenAI)
- Offers integrations: API docs, LangChain, Llama Index, Pipedream, Zapier, n8n, Dify, Toolhouse, MCP server
Recent releases
View all 16 releases →
v2.0.0
Breaking risk
Breaking changes
- Requires scrapegraph-py v2.0.0 or later
- API surface aligned with scrapegraph-py v2
- Minimum Python version raised to 3.12
Full changelog
2.0.0 (2026-04-19)
⚠ BREAKING CHANGES
- requires scrapegraph-py v2.0.0+
Co-Authored-By: Claude Opus 4.6 (1M context) [email protected]
Features
- add scrapegraph-py PR #84 SDK compatibility (e8b2a28), closes #82
- align with scrapegraph-py v2 API surface from PR #82 (c0f5fd5)
- migrate to scrapegraph-py v2 API surface (fd23bb0), closes ScrapeGraphAI/scrapegraph-py#82
CI
- bump min Python to 3.12 and trim test suite (5fda03f)
v1.76.0
New feature
Notable features
- PlasmateLoader: lightweight scraping backend with no Chrome dependency
Full changelog
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
About
Stars
28,463
Forks
2,767
Languages
Python
Makefile
Dockerfile
Install & Platforms
Install via
pip
