9 projects
FastAPI is a high-performance Python web framework for building APIs that leverages standard type hints, Starlette for the web layer, and Pydantic for data validation. It automatically generates OpenAPI-based interactive docs and provides editor autocompletion and request/response validation from type annotations.
Flask is a lightweight Python WSGI web framework built on Werkzeug and Jinja that provides a minimal core with routing, templating, and request handling, leaving architectural decisions and extra functionality to the developer and a large extension ecosystem.
Scrapling is a Python web scraping framework whose parser can "adaptively" relocate elements when a site's HTML changes, paired with stealth fetchers that bypass anti-bot systems like Cloudflare Turnstile and a Scrapy-style spider engine for concurrent, resumable, proxy-rotating crawls.
Scrapy is a mature Python framework for building web crawlers and scrapers, offering an asynchronous engine, built-in support for following links, extracting structured data via selectors, and exporting results through configurable pipelines.
Quivr is an opinionated Python framework for building Retrieval-Augmented Generation (RAG) pipelines, letting developers ingest documents into a vector store and query them with LLMs. It packages chunking, embedding, retrieval, and answer generation behind a configurable "brain" abstraction.
Textual is a Python framework for building interactive applications with a CSS-styled, widget-based, reactive API that run in the terminal and can also be served to a web browser. It brings web-style layout and theming to TUIs while staying async-optional.
Composio is an integration platform that gives AI agents and LLMs authenticated access to hundreds of external tools and apps (Gmail, Slack, GitHub, etc.) through a managed SDK, handling OAuth, auth config, and tool execution so agents can take real-world actions.
Haystack is an open-source Python framework for orchestrating LLM applications through modular, traceable pipelines and agent workflows. It gives explicit control over retrieval, routing, memory, and generation to build RAG systems, semantic search, and autonomous agents that are model- and vendor-agnostic.
MemMachine is an open-source long-term memory layer for AI agents that stores and recalls information across sessions using episodic (graph-based), profile (SQL), and working memory types. It exposes Python/TypeScript SDKs, a REST API, and an MCP server, and integrates with frameworks like LangChain, LangGraph, and CrewAI.