feat: use explicit Projects section in README

2026-05-06 06:06:45 -04:00 · 2026-05-04 21:24:57 +08:00
parent 921d47b455
commit 6c18b6447e
5 changed files with 190 additions and 178 deletions
@@ -2,13 +2,13 @@
 An opinionated guide to the best Python frameworks, libraries, tools, and resources.
-# **Sponsors**
+## **Sponsors**
 - **[pyr](https://pyrun.dev)**: Zero-config Python project manager. Bootstraps its own runtime, app-convention, and working imports - out the box.
 > The **#10 most-starred repo on GitHub**. Put your product in front of Python developers. [Become a sponsor](SPONSORSHIP.md).
-# Categories
+## Categories
 **AI & ML**
@@ -123,11 +123,11 @@ An opinionated guide to the best Python frameworks, libraries, tools, and resour
 - [Microsoft Windows](#microsoft-windows)
 - [Miscellaneous](#miscellaneous)
---
+## Projects
 **AI & ML**
-## AI and Agents
+### AI and Agents
 _Libraries for building AI applications, LLM integrations, and autonomous agents._
@@ -163,7 +163,7 @@ _Libraries for building AI applications, LLM integrations, and autonomous agents
  - [vibevoice](https://github.com/microsoft/VibeVoice) - A family of open-source voice AI models from Microsoft for text-to-speech and long-form speech recognition.
  - [voxcpm](https://github.com/OpenBMB/VoxCPM) - A tokenizer-free text-to-speech foundation model for multilingual speech generation and voice cloning.
-## Deep Learning
+### Deep Learning
 _Frameworks for Neural Networks and Deep Learning. Also see [awesome-deep-learning](https://github.com/ChristosChristofidis/awesome-deep-learning)._
@@ -174,7 +174,7 @@ _Frameworks for Neural Networks and Deep Learning. Also see [awesome-deep-learni
 - [stable-baselines3](https://github.com/DLR-RM/stable-baselines3) - PyTorch implementations of Stable Baselines (deep) reinforcement learning algorithms.
 - [tensorflow](https://github.com/tensorflow/tensorflow) - The most popular Deep Learning framework created by Google.
-## Machine Learning
+### Machine Learning
 _Libraries for Machine Learning. Also see [awesome-machine-learning](https://github.com/josephmisiti/awesome-machine-learning#python)._
@@ -190,7 +190,7 @@ _Libraries for Machine Learning. Also see [awesome-machine-learning](https://git
 - [timesfm](https://github.com/google-research/timesfm) - A pretrained foundation model from Google Research for time-series forecasting.
 - [xgboost](https://github.com/dmlc/xgboost) - A scalable, portable, and distributed gradient boosting library.
-## Natural Language Processing
+### Natural Language Processing
 _Libraries for working with human languages._
@@ -203,7 +203,7 @@ _Libraries for working with human languages._
  - [funnlp](https://github.com/fighting41love/funNLP) - A collection of tools and datasets for Chinese NLP.
  - [jieba](https://github.com/fxsjy/jieba) - The most popular Chinese text segmentation library.
-## Computer Vision
+### Computer Vision
 _Libraries for Computer Vision._
@@ -212,7 +212,7 @@ _Libraries for Computer Vision._
 - [opencv](https://github.com/opencv/opencv-python) - Open Source Computer Vision Library.
 - [pytesseract](https://github.com/madmaze/pytesseract) - A wrapper for [Google Tesseract OCR](https://github.com/tesseract-ocr).
-## Recommender Systems
+### Recommender Systems
 _Libraries for building recommender systems._
@@ -222,7 +222,7 @@ _Libraries for building recommender systems._
 **Web Development**
-## Web Frameworks
+### Web Frameworks
 _Traditional full stack web frameworks. Also see [Web APIs](#web-apis)._
@@ -245,7 +245,7 @@ _Traditional full stack web frameworks. Also see [Web APIs](#web-apis)._
  - [starlette](https://github.com/Kludex/starlette) - A lightweight ASGI framework and toolkit for building high-performance async services.
  - [tornado](https://github.com/tornadoweb/tornado) - A web framework and asynchronous networking library.
-## Web APIs
+### Web APIs
 _Libraries for building RESTful and GraphQL APIs._
@@ -264,7 +264,7 @@ _Libraries for building RESTful and GraphQL APIs._
  - [strawberry](https://github.com/strawberry-graphql/strawberry) - A GraphQL library that leverages Python type annotations for schema definition.
  - [webargs](https://github.com/marshmallow-code/webargs) - A friendly library for parsing HTTP request arguments with built-in support for popular web frameworks.
-## Web Servers
+### Web Servers
 _ASGI and WSGI compatible web servers._
@@ -281,7 +281,7 @@ _ASGI and WSGI compatible web servers._
  - [grpcio](https://github.com/grpc/grpc) - HTTP/2-based RPC framework with Python bindings, built by Google.
  - [rpyc](https://github.com/tomerfiliba-org/rpyc) (Remote Python Call) - A transparent and symmetric RPC library for Python.
-## WebSocket
+### WebSocket
 _Libraries for working with WebSocket._
@@ -291,21 +291,21 @@ _Libraries for working with WebSocket._
 - [picows](https://github.com/tarasko/picows) - Fastest WebSocket clients and servers with a frame level interface for the most demanding use-cases.
 - [websockets](https://github.com/python-websockets/websockets) - A library for building WebSocket servers and clients with a focus on correctness and simplicity.
-## Template Engines
+### Template Engines
 _Libraries and tools for templating and lexing._
 - [jinja](https://github.com/pallets/jinja) - A modern and designer friendly templating language.
 - [mako](https://github.com/sqlalchemy/mako) - Hyperfast and lightweight templating for the Python platform.
-## Web Asset Management
+### Web Asset Management
 _Tools for managing, compressing and minifying website assets._
 - [django-compressor](https://github.com/django-compressor/django-compressor) - Compresses linked and inline JavaScript or CSS into a single cached file.
 - [django-storages](https://github.com/jschneier/django-storages) - A collection of custom storage back ends for Django.
-## Authentication
+### Authentication
 _Libraries for implementing authentication schemes._
@@ -320,7 +320,7 @@ _Libraries for implementing authentication schemes._
  - [django-guardian](https://github.com/django-guardian/django-guardian) - Implementation of per object permissions for Django 1.2+
  - [django-rules](https://github.com/dfunckt/django-rules) - A tiny but powerful app providing object-level permissions to Django, without requiring a database.
-## Admin Panels
+### Admin Panels
 _Libraries for administrative interfaces._
@@ -332,7 +332,7 @@ _Libraries for administrative interfaces._
 - [func-to-web](https://github.com/offerrall/FuncToWeb) - Instantly create web UIs from Python functions using type hints. Zero frontend code required.
 - [jet-bridge](https://github.com/jet-admin/jet-bridge) - Admin panel framework for any application with nice UI (ex Jet Django).
-## CMS
+### CMS
 _Content Management Systems._
@@ -340,7 +340,7 @@ _Content Management Systems._
 - [indico](https://github.com/indico/indico) - A feature-rich event management system, made @ [CERN](https://en.wikipedia.org/wiki/CERN).
 - [wagtail](https://github.com/wagtail/wagtail) - A Django content management system.
-## Static Site Generators
+### Static Site Generators
 _Static site generator is a software that takes some text + templates as input and produces HTML files on the output._
@@ -350,7 +350,7 @@ _Static site generator is a software that takes some text + templates as input a
 **HTTP & Scraping**
-## HTTP Clients
+### HTTP Clients
 _Libraries for working with HTTP._
@@ -361,7 +361,7 @@ _Libraries for working with HTTP._
 - [requests](https://github.com/psf/requests) - HTTP Requests for Humans.
 - [urllib3](https://github.com/urllib3/urllib3) - A HTTP library with thread-safe connection pooling, file post support, sanity friendly.
-## Web Scraping
+### Web Scraping
 _Libraries to automate web scraping and extract web content._
@@ -377,7 +377,7 @@ _Libraries to automate web scraping and extract web content._
  - [sumy](https://github.com/miso-belica/sumy) - A module for automatic summarization of text documents and HTML pages.
  - [trafilatura](https://github.com/adbar/trafilatura) - A tool for gathering text and metadata from the web, with built-in content filtering.
-## Email
+### Email
 _Libraries for sending and parsing email, and mail server management._
@@ -386,7 +386,7 @@ _Libraries for sending and parsing email, and mail server management._
 **Database & Storage**
-## ORM
+### ORM
 _Libraries that implement Object-Relational Mapping or data mapping techniques._
@@ -404,7 +404,7 @@ _Libraries that implement Object-Relational Mapping or data mapping techniques._
  - [mongoengine](https://github.com/MongoEngine/mongoengine) - A Python Object-Document-Mapper for working with MongoDB.
  - [pynamodb](https://github.com/pynamodb/PynamoDB) - A Pythonic interface for [Amazon DynamoDB](https://aws.amazon.com/dynamodb/).
-## Database Drivers
+### Database Drivers
 _Libraries for connecting and operating databases._
@@ -425,7 +425,7 @@ _Libraries for connecting and operating databases._
  - [pymongo](https://github.com/mongodb/mongo-python-driver) - The official Python client for MongoDB.
  - [redis-py](https://github.com/redis/redis-py) - The Python client for Redis.
-## Database
+### Database
 _Databases implemented in Python._
@@ -435,7 +435,7 @@ _Databases implemented in Python._
 - [tinydb](https://github.com/msiemens/tinydb) - A tiny, document-oriented database.
 - [ZODB](https://github.com/zopefoundation/ZODB) - A native object database for Python. A key-value and object graph database.
-## Caching
+### Caching
 _Libraries for caching data._
@@ -444,7 +444,7 @@ _Libraries for caching data._
 - [dogpile.cache](https://github.com/sqlalchemy/dogpile.cache) - dogpile.cache is a next generation replacement for Beaker made by the same authors.
 - [python-diskcache](https://github.com/grantjenks/python-diskcache) - SQLite and file backed cache backend with faster lookups than memcached and redis.
-## Search
+### Search
 _Libraries and software for indexing and performing search queries on data._
@@ -452,7 +452,7 @@ _Libraries and software for indexing and performing search queries on data._
 - [elasticsearch-py](https://github.com/elastic/elasticsearch-py) - The official low-level Python client for [Elasticsearch](https://www.elastic.co/products/elasticsearch).
 - [pysolr](https://github.com/django-haystack/pysolr) - A lightweight Python wrapper for [Apache Solr](https://lucene.apache.org/solr/).
-## Serialization
+### Serialization
 _Libraries for serializing complex data types._
@@ -462,7 +462,7 @@ _Libraries for serializing complex data types._
 **Data & Science**
-## Data Analysis
+### Data Analysis
 _Libraries for data analysis._
@@ -483,7 +483,7 @@ _Libraries for data analysis._
  - [openbb](https://github.com/OpenBB-finance/OpenBB) - A financial data platform for analysts, quants and AI agents.
  - [yfinance](https://github.com/ranaroussi/yfinance) - Easy Pythonic way to download market and financial data from Yahoo Finance.
-## Data Validation
+### Data Validation
 _Libraries for validating data. Used for forms in many cases._
@@ -493,7 +493,7 @@ _Libraries for validating data. Used for forms in many cases._
 - [pydantic](https://github.com/pydantic/pydantic) - Data validation using Python type hints.
 - [voluptuous](https://github.com/alecthomas/voluptuous) - A Python data validation library primarily intended for validating data from untrusted sources.
-## Data Visualization
+### Data Visualization
 _Libraries for visualizing data. Also see [awesome-javascript](https://github.com/sorrycc/awesome-javascript#data-visualization)._
@@ -516,7 +516,7 @@ _Libraries for visualizing data. Also see [awesome-javascript](https://github.co
  - [gradio](https://github.com/gradio-app/gradio) - Build and share machine learning apps, all in Python.
  - [streamlit](https://github.com/streamlit/streamlit) - A framework which lets you build dashboards, generate reports, or create chat apps in minutes.
-## Geolocation
+### Geolocation
 _Libraries for geocoding addresses and working with latitudes and longitudes._
@@ -526,7 +526,7 @@ _Libraries for geocoding addresses and working with latitudes and longitudes._
 - [geopandas](https://github.com/geopandas/geopandas) - Python tools for geographic data (GeoSeries/GeoDataFrame) built on pandas.
 - [geopy](https://github.com/geopy/geopy) - Python Geocoding Toolbox.
-## Science
+### Science
 _Libraries for scientific computing. Also see [Python-for-Scientists](https://github.com/TomNicholas/Python-for-Scientists)._
@@ -557,7 +557,7 @@ _Libraries for scientific computing. Also see [Python-for-Scientists](https://gi
  - [networkx](https://github.com/networkx/networkx) - A high-productivity software for complex networks.
  - [shapely](https://github.com/shapely/shapely) - Manipulation and analysis of geometric objects in the Cartesian plane.
-## Quantum Computing
+### Quantum Computing
 _Libraries for quantum computing._
@@ -568,7 +568,7 @@ _Libraries for quantum computing._
 **Developer Tools**
-## Algorithms and Design Patterns
+### Algorithms and Design Patterns
 _Python implementation of data structures, algorithms and design patterns. Also see [awesome-algorithms](https://github.com/tayllan/awesome-algorithms)._
@@ -580,7 +580,7 @@ _Python implementation of data structures, algorithms and design patterns. Also
  - [python-patterns](https://github.com/faif/python-patterns) - A collection of design patterns in Python.
  - [transitions](https://github.com/pytransitions/transitions) - A lightweight, object-oriented finite state machine implementation.
-## Interactive Interpreter
+### Interactive Interpreter
 _Interactive Python interpreters (REPL)._
@@ -589,7 +589,7 @@ _Interactive Python interpreters (REPL)._
 - [marimo](https://github.com/marimo-team/marimo) - Transform data and train models, feels like a next-gen notebook, stored as Git-friendly Python.
 - [ptpython](https://github.com/prompt-toolkit/ptpython) - Advanced Python REPL built on top of the [python-prompt-toolkit](https://github.com/prompt-toolkit/python-prompt-toolkit).
-## Code Analysis
+### Code Analysis
 _Tools of static analysis, linters and code quality checkers. Also see [awesome-static-analysis](https://github.com/analysis-tools-dev/static-analysis)._
@@ -619,7 +619,7 @@ _Tools of static analysis, linters and code quality checkers. Also see [awesome-
  - [monkeytype](https://github.com/Instagram/MonkeyType) - A system for Python that generates static type annotations by collecting runtime types.
  - [pytype](https://github.com/google/pytype) - Pytype checks and infers types for Python code - without requiring type annotations.
-## Testing
+### Testing
 _Libraries for testing codebases and generating test data. Also see [awesome-python-testing](https://github.com/cleder/awesome-python-testing)._
@@ -655,7 +655,7 @@ _Libraries for testing codebases and generating test data. Also see [awesome-pyt
  - [faker](https://github.com/joke2k/faker) - A Python package that generates fake data.
  - [mimesis](https://github.com/lk-geimfari/mimesis) - is a Python library that help you generate fake data.
-## Debugging Tools
+### Debugging Tools
 _Libraries for debugging code._
@@ -674,7 +674,7 @@ _Libraries for debugging code._
  - [icecream](https://github.com/gruns/icecream) - Inspect variables, expressions, and program execution with a single, simple function call.
  - [memory_graph](https://github.com/bterwijn/memory_graph) - Visualize Python data at runtime to debug references, mutability, and aliasing.
-## Build Tools
+### Build Tools
 _Compile software from source code._
@@ -685,7 +685,7 @@ _Compile software from source code._
 - [doit](https://github.com/pydoit/doit) - A task runner and build tool.
 - [scons](https://github.com/SCons/scons) - A software construction tool.
-## Documentation
+### Documentation
 _Libraries for generating project documentation._
@@ -697,7 +697,7 @@ _Libraries for generating project documentation._
 **DevOps**
-## DevOps Tools
+### DevOps Tools
 _Software and libraries for DevOps._
@@ -723,7 +723,7 @@ _Software and libraries for DevOps._
  - [chaostoolkit](https://github.com/chaostoolkit/chaostoolkit) - A Chaos Engineering toolkit & Orchestration for Developers.
  - [pre-commit](https://github.com/pre-commit/pre-commit) - A framework for managing and maintaining multi-language pre-commit hooks.
-## Distributed Computing
+### Distributed Computing
 _Frameworks and libraries for Distributed Computing._
@@ -735,7 +735,7 @@ _Frameworks and libraries for Distributed Computing._
  - [joblib](https://github.com/joblib/joblib) - A set of tools to provide lightweight pipelining in Python.
  - [ray](https://github.com/ray-project/ray/) - A system for parallel and distributed Python that unifies the machine learning ecosystem.
-## Task Queues
+### Task Queues
 _Libraries for working with task queues._
@@ -744,13 +744,13 @@ _Libraries for working with task queues._
 - [huey](https://github.com/coleifer/huey) - Little multi-threaded task queue.
 - [rq](https://github.com/rq/rq) - Simple job queues for Python.
-## Messaging
+### Messaging
 _Libraries for working with message brokers and event streaming._
 - [faststream](https://github.com/ag2ai/faststream) - A framework for building asynchronous services over Apache Kafka, RabbitMQ, NATS, MQTT and Redis.
-## Job Schedulers
+### Job Schedulers
 _Libraries for scheduling jobs._
@@ -761,7 +761,7 @@ _Libraries for scheduling jobs._
 - [schedule](https://github.com/dbader/schedule) - Python job scheduling for humans.
 - [SpiffWorkflow](https://github.com/sartography/SpiffWorkflow) - A powerful workflow engine implemented in pure Python.
-## Logging
+### Logging
 _Libraries for generating and working with logs._
@@ -770,7 +770,7 @@ _Libraries for generating and working with logs._
 - [loguru](https://github.com/Delgan/loguru) - Library which aims to bring enjoyable logging in Python.
 - [structlog](https://github.com/hynek/structlog) - Structured logging made easy.
-## Network Virtualization
+### Network Virtualization
 _Tools and libraries for Virtual Networking and SDN (Software Defined Networking)._
@@ -780,7 +780,7 @@ _Tools and libraries for Virtual Networking and SDN (Software Defined Networking
 **CLI & GUI**
-## CLI Development
+### CLI Development
 _Libraries for building command-line applications._
@@ -799,7 +799,7 @@ _Libraries for building command-line applications._
  - [textual](https://github.com/Textualize/textual) - A framework for building interactive user interfaces that run in the terminal and the browser.
  - [tqdm](https://github.com/tqdm/tqdm) - Fast, extensible progress bar for loops and CLI.
-## CLI Tools
+### CLI Tools
 _Useful CLI-based tools for productivity._
@@ -818,7 +818,7 @@ _Useful CLI-based tools for productivity._
  - [mycli](https://github.com/dbcli/mycli) - MySQL CLI with autocompletion and syntax highlighting.
  - [pgcli](https://github.com/dbcli/pgcli) - PostgreSQL CLI with autocompletion and syntax highlighting.
-## GUI Development
+### GUI Development
 _Libraries for working with graphical user interface applications._
@@ -846,7 +846,7 @@ _Libraries for working with graphical user interface applications._
 **Text & Documents**
-## Text Processing
+### Text Processing
 _Libraries for parsing and manipulating plain texts._
@@ -872,7 +872,7 @@ _Libraries for parsing and manipulating plain texts._
  - [python-user-agents](https://github.com/selwin/python-user-agents) - Browser user agent parser.
  - [sqlparse](https://github.com/andialbrecht/sqlparse) - A non-validating SQL parser.
-## HTML Manipulation
+### HTML Manipulation
 _Libraries for working with HTML and XML._
@@ -884,7 +884,7 @@ _Libraries for working with HTML and XML._
 - [tinycss2](https://github.com/Kozea/tinycss2) - A low-level CSS parser and generator written in Python.
 - [xmltodict](https://github.com/martinblech/xmltodict) - Working with XML feel like you are working with JSON.
-## File Format Processing
+### File Format Processing
 _Libraries for parsing and manipulating specific text formats._
@@ -918,7 +918,7 @@ _Libraries for parsing and manipulating specific text formats._
  - [pyyaml](https://github.com/yaml/pyyaml) - YAML implementations for Python.
  - [tomllib](https://docs.python.org/3/library/tomllib.html) - (Python standard library) Parse TOML files.
-## File Manipulation
+### File Manipulation
 _Libraries for file manipulation._
@@ -930,7 +930,7 @@ _Libraries for file manipulation._
 **Media**
-## Image Processing
+### Image Processing
 _Libraries for manipulating images._
@@ -943,7 +943,7 @@ _Libraries for manipulating images._
 - [thumbor](https://github.com/thumbor/thumbor) - A smart imaging service. It enables on-demand crop, re-sizing and flipping of images.
 - [wand](https://github.com/emcconville/wand) - Python bindings for [MagickWand](https://www.imagemagick.org/script/magick-wand.php), C API for ImageMagick.
-## Audio & Video Processing
+### Audio & Video Processing
 _Libraries for manipulating audio, video, and their metadata._
@@ -960,7 +960,7 @@ _Libraries for manipulating audio, video, and their metadata._
  - [mutagen](https://github.com/quodlibet/mutagen) - A Python module to handle audio metadata.
  - [tinytag](https://github.com/devsnd/tinytag) - A library for reading music meta data of MP3, OGG, FLAC and Wave files.
-## Game Development
+### Game Development
 _Awesome game development libraries._
@@ -973,7 +973,7 @@ _Awesome game development libraries._
 **Python Language**
-## Implementations
+### Implementations
 _Implementations of Python._
@@ -984,7 +984,7 @@ _Implementations of Python._
 - [pyodide](https://github.com/pyodide/pyodide) - Python distribution for the browser and Node.js based on WebAssembly.
 - [pypy](https://github.com/pypy/pypy) - A very fast and compliant implementation of the Python language.
-## Built-in Classes Enhancement
+### Built-in Classes Enhancement
 _Libraries for enhancing Python built-in classes._
@@ -992,7 +992,7 @@ _Libraries for enhancing Python built-in classes._
 - [bidict](https://github.com/jab/bidict) - Efficient, Pythonic bidirectional map data structures and related functionality.
 - [box](https://github.com/cdgriffith/Box) - Python dictionaries with advanced dot notation access.
-## Functional Programming
+### Functional Programming
 _Functional Programming with Python._
@@ -1003,7 +1003,7 @@ _Functional Programming with Python._
 - [returns](https://github.com/dry-python/returns) - A set of type-safe monads, transformers, and composition utilities.
 - [toolz](https://github.com/pytoolz/toolz) - A collection of functional utilities for iterators, functions, and dictionaries. Also available as [cytoolz](https://github.com/pytoolz/cytoolz/) for Cython-accelerated performance.
-## Asynchronous Programming
+### Asynchronous Programming
 _Libraries for asynchronous, concurrent and parallel execution. Also see [awesome-asyncio](https://github.com/timofurrer/awesome-asyncio)._
@@ -1017,7 +1017,7 @@ _Libraries for asynchronous, concurrent and parallel execution. Also see [awesom
 - [twisted](https://github.com/twisted/twisted) - An event-driven networking engine.
 - [uvloop](https://github.com/MagicStack/uvloop) - Ultra fast asyncio event loop.
-## Date and Time
+### Date and Time
 _Libraries for working with dates and times._
@@ -1028,7 +1028,7 @@ _Libraries for working with dates and times._
 **Python Toolchain**
-## Environment Management
+### Environment Management
 _Libraries for Python version and virtual environment management._
@@ -1038,7 +1038,7 @@ _Libraries for Python version and virtual environment management._
 - [uv](https://github.com/astral-sh/uv) - An extremely fast Python version, package and project manager, written in Rust.
 - [virtualenv](https://github.com/pypa/virtualenv) - A tool to create isolated Python environments.
-## Package Management
+### Package Management
 _Libraries for package and dependency management._
@@ -1048,7 +1048,7 @@ _Libraries for package and dependency management._
 - [poetry](https://github.com/python-poetry/poetry) - Python dependency management and packaging made easy.
 - [uv](https://github.com/astral-sh/uv) - An extremely fast Python version, package and project manager, written in Rust.
-## Package Repositories
+### Package Repositories
 _Local PyPI repository server and proxies._
@@ -1056,7 +1056,7 @@ _Local PyPI repository server and proxies._
 - [devpi](https://github.com/devpi/devpi) - PyPI server and packaging/testing/release tool.
 - [warehouse](https://github.com/pypa/warehouse) - Next generation Python Package Repository (PyPI).
-## Distribution
+### Distribution
 _Libraries to create packaged executables for release distribution._
@@ -1066,7 +1066,7 @@ _Libraries to create packaged executables for release distribution._
 - [pyinstaller](https://github.com/pyinstaller/pyinstaller) - Converts Python programs into stand-alone executables (cross-platform).
 - [shiv](https://github.com/linkedin/shiv) - A command line utility for building fully self-contained zipapps (PEP 441), but with all their dependencies included.
-## Configuration Files
+### Configuration Files
 _Libraries for storing and parsing configuration options._
@@ -1078,13 +1078,13 @@ _Libraries for storing and parsing configuration options._
 **Security**
-## Cryptography
+### Cryptography
 - [cryptography](https://github.com/pyca/cryptography) - A package designed to expose cryptographic primitives and recipes to Python developers.
 - [paramiko](https://github.com/paramiko/paramiko) - The leading native Python SSHv2 protocol library.
 - [pynacl](https://github.com/pyca/pynacl) - Python binding to the Networking and Cryptography (NaCl) library.
-## Penetration Testing
+### Penetration Testing
 _Frameworks and tools for penetration testing._
@@ -1093,7 +1093,7 @@ _Frameworks and tools for penetration testing._
 - [sherlock](https://github.com/sherlock-project/sherlock) - Hunt down social media accounts by username across social networks.
 - [sqlmap](https://github.com/sqlmapproject/sqlmap) - Automatic SQL injection and database takeover tool.
-## Web Security
+### Web Security
 _Libraries for application-layer web security._
@@ -1101,14 +1101,14 @@ _Libraries for application-layer web security._
 **Other**
-## Hardware
+### Hardware
 _Libraries for programming with hardware._
 - [bleak](https://github.com/hbldh/bleak) - A cross platform Bluetooth Low Energy Client for Python using asyncio.
 - [pynput](https://github.com/moses-palmer/pynput) - A library to control and monitor input devices.
-## Microsoft Windows
+### Microsoft Windows
 _Python programming on Microsoft Windows._
@@ -1116,7 +1116,7 @@ _Python programming on Microsoft Windows._
 - [pywin32](https://github.com/mhammond/pywin32) - Python Extensions for Windows.
 - [winpython](https://github.com/winpython/winpython) - Portable development environment for Windows 10/11.
-## Miscellaneous
+### Miscellaneous
 _Useful libraries or tools that don't fit in the categories above._
@@ -1125,18 +1125,18 @@ _Useful libraries or tools that don't fit in the categories above._
 - [itsdangerous](https://github.com/pallets/itsdangerous) - Various helpers to pass trusted data to untrusted environments.
 - [tryton](https://github.com/tryton/tryton) - A general-purpose business framework.
-# Resources
+## Resources
 Where to discover learning resources or new Python libraries.
-## Newsletters
+### Newsletters
 - [Awesome Python Newsletter](http://python.libhunt.com/newsletter)
 - [Pycoder's Weekly](https://pycoders.com/)
 - [Python Tricks](https://realpython.com/python-tricks/)
 - [Python Weekly](https://www.pythonweekly.com/)
-## Podcasts
+### Podcasts
 - [Django Chat](https://djangochat.com/)
 - [PyPodcats](https://pypodcats.live)
@@ -1144,11 +1144,11 @@ Where to discover learning resources or new Python libraries.
 - [Talk Python To Me](https://talkpython.fm/)
 - [The Real Python Podcast](https://realpython.com/podcasts/rpp/)
-## Websites
+### Websites
 - [Python Developer Tooling Handbook](https://pydevtools.com/) - Comprehensive guide to modern Python developer tools covering package management, linting, type checking, testing, and more.
-# Contributing
+## Contributing
 Your contributions are always welcome! Please take a look at the [contribution guidelines](https://github.com/vinta/awesome-python/blob/master/CONTRIBUTING.md) first.
@@ -243,13 +243,14 @@ def write_sitemap_xml(path: Path, urls: Sequence[tuple[str, str]]) -> None:
 def top_level_heading_text(line: str) -> str | None:
    stripped = line.strip()
-    if not stripped.startswith("# "):
+    match = re.match(r"^(#{1,2})\s+(.+)$", stripped)
    if match is None:
        return None
-    return stripped.removeprefix("#").strip().strip("#").strip().strip("*").strip()
+    return match.group(2).strip().strip("#").strip().strip("*").strip()
 def extract_categories_body(markdown: str) -> str:
-    """Return content under the `# Categories` heading, excluding the heading line itself."""
+    """Return content from `Categories` through `Projects`, excluding later sections."""
    lines = markdown.splitlines(keepends=True)
    start_idx = None
    end_idx = len(lines)
@@ -261,7 +262,7 @@ def extract_categories_body(markdown: str) -> str:
            start_idx = i + 1
            while start_idx < len(lines) and lines[start_idx].strip() == "":
                start_idx += 1
-        elif start_idx is not None and i >= start_idx:
+        elif start_idx is not None and heading.lower() in ("resources", "contributing"):
            end_idx = i
            break
    if start_idx is None:
@@ -114,6 +114,13 @@ def _heading_text(node: SyntaxTreeNode) -> str:
    return ""
 def _heading_level(node: SyntaxTreeNode) -> int | None:
    """Return the numeric level for a heading node."""
    if node.type != "heading" or not node.tag.startswith("h"):
        return None
    return int(node.tag[1:])
 def _extract_description_children(nodes: list[SyntaxTreeNode]) -> list[SyntaxTreeNode]:
    """Extract description children from the first paragraph if it's a single <em> block.
@@ -303,7 +310,7 @@ def _parse_grouped_sections(
 ) -> list[ParsedGroup]:
    """Parse nodes into groups of categories using bold markers as group boundaries.
-    Bold-only paragraphs (**Group Name**) delimit groups. H2 headings under each
+    Bold-only paragraphs (**Group Name**) delimit groups. H3 headings under each
    bold marker become categories within that group. Categories appearing before
    any bold marker go into an "Other" group.
    """
@@ -341,7 +348,7 @@ def _parse_grouped_sections(
            flush_group()
            current_group_name = bold_name
            current_cat_body = []
-        elif node.type == "heading" and node.tag == "h2":
+        elif node.type == "heading" and node.tag in ("h2", "h3"):
            flush_cat()
            current_cat_name = _heading_text(node)
            current_cat_body = []
@@ -383,7 +390,7 @@ def _parse_sponsor_item(inline: SyntaxTreeNode) -> ParsedSponsor | None:
 def parse_sponsors(text: str) -> list[ParsedSponsor]:
-    """Parse the `# Sponsors` section of README.md into a list of sponsors.
+    """Parse the `Sponsors` section of README.md into a list of sponsors.
    Expects bullets in the form `**[name](url)**: description`.
    Returns [] if no Sponsors section exists.
@@ -395,14 +402,18 @@ def parse_sponsors(text: str) -> list[ParsedSponsor]:
    start_idx = None
    end_idx = len(children)
    start_level = None
    for i, node in enumerate(children):
-        if node.type == "heading" and node.tag == "h1":
+        level = _heading_level(node)
-            title = _heading_text(node).strip().lower()
+        if level is None:
-            if start_idx is None and title == "sponsors":
+            continue
-                start_idx = i + 1
+        title = _heading_text(node).strip().lower()
-            elif start_idx is not None:
+        if start_idx is None and title == "sponsors":
-                end_idx = i
+            start_idx = i + 1
-                break
+            start_level = level
        elif start_idx is not None and start_level is not None and level <= start_level:
            end_idx = i
            break
    if start_idx is None:
        return []
@@ -426,26 +437,26 @@ def parse_readme(text: str) -> list[ParsedGroup]:
    """Parse README.md text into grouped categories.
    Returns a list of ParsedGroup dicts containing nested categories.
-    Content between the thematic break (---) and # Resources or # Contributing
+    Content between the Projects heading and Resources or Contributing is parsed
-    is parsed as categories grouped by bold markers (**Group Name**).
+    as categories grouped by bold markers (**Group Name**).
    """
    md = MarkdownIt("commonmark")
    tokens = md.parse(text)
    root = SyntaxTreeNode(tokens)
    children = root.children
-    # Find thematic break (---) and section boundaries in one pass
+    # Find Projects and section boundaries in one pass.
-    hr_idx = None
+    projects_idx = None
    cat_end_idx = None
    for i, node in enumerate(children):
-        if hr_idx is None and node.type == "hr":
+        if _heading_level(node) in (1, 2):
            hr_idx = i
        elif node.type == "heading" and node.tag == "h1":
            text_content = _heading_text(node)
-            if cat_end_idx is None and text_content in ("Resources", "Contributing"):
+            if projects_idx is None and text_content == "Projects":
                projects_idx = i
            elif cat_end_idx is None and text_content in ("Resources", "Contributing"):
                cat_end_idx = i
-    if hr_idx is None:
+    if projects_idx is None:
        return []
-    cat_nodes = children[hr_idx + 1 : cat_end_idx or len(children)]
+    cat_nodes = children[projects_idx + 1 : cat_end_idx or len(children)]
    return _parse_grouped_sections(cat_nodes)
@@ -137,31 +137,31 @@ class TestBuild:
            Intro.
-            ---
+            ## Projects
            **Tools**
-            ## Widgets
+            ### Widgets
            _Widget libraries. Also see [awesome-widgets](https://example.com/widgets)._
            - [w1](https://example.com) - A widget.
-            ## Gadgets
+            ### Gadgets
            _Gadget tools._
            - [g1](https://example.com) - A gadget.
-            # Resources
+            ## Resources
            Info.
-            ## Newsletters
+            ### Newsletters
            - [NL](https://example.com)
-            # Contributing
+            ## Contributing
            Help!
        """)
@@ -179,17 +179,17 @@ class TestBuild:
            Intro.
-            ---
+            ## Projects
            **Tools**
-            ## Widgets
+            ### Widgets
            - Sync
                - [w1](https://example.com) - A widget.
-            # Contributing
+            ## Contributing
            Help!
        """)
@@ -232,7 +232,7 @@ class TestBuild:
            Intro.
-            ---
+            ## Projects
            **Tools**
@@ -298,28 +298,28 @@ class TestBuild:
            Intro.
-            # **Sponsors**
+            ## **Sponsors**
            - **[Sponsor](https://sponsor.example.com)**: Sponsored tool.
            > Become a sponsor: [Sponsor us](SPONSORSHIP.md).
-            # Categories
+            ## Categories
            **Tools**
            - [Widgets](#widgets)
-            ---
+            ## Projects
            **Tools**
-            ## Widgets
+            ### Widgets
            - [w1](https://example.com) - A widget.
            - [w2](https://github.com/owner/w2) - A starred widget.
-            # Contributing
+            ## Contributing
            Help!
        """)
@@ -353,7 +353,7 @@ class TestBuild:
        assert "## Categories" in llms_txt
        assert "**Tools**" in llms_txt
        assert "- [Widgets](#widgets)" in llms_txt
-        assert "## Widgets" in llms_txt
+        assert "### Widgets" in llms_txt
        assert "- [w1](https://example.com) - A widget." in llms_txt
        assert "- [w2](https://github.com/owner/w2) - A starred widget. (GitHub stars: 42)" in llms_txt
        assert llms_txt != readme
@@ -363,7 +363,7 @@ class TestBuild:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            ## Only
@@ -387,7 +387,7 @@ class TestBuild:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            ## Stuff
@@ -431,7 +431,7 @@ class TestBuild:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **Widgets**
@@ -538,7 +538,7 @@ class TestBuild:
            Intro.
-            ---
+            ## Projects
            **Tools**
@@ -591,7 +591,7 @@ class TestBuild:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **AI & ML**
@@ -624,7 +624,7 @@ class TestBuild:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **Web**
@@ -659,7 +659,7 @@ class TestBuild:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **Web**
@@ -691,7 +691,7 @@ class TestBuild:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **AI & ML**
@@ -731,7 +731,7 @@ class TestBuild:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            ## Sneaky </script><script>x=1</script>
@@ -760,7 +760,7 @@ class TestBuild:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **AI & ML**
@@ -800,7 +800,7 @@ class TestBuild:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **AI & ML**
@@ -980,7 +980,7 @@ class TestExtractEntries:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **Tools**
@@ -1004,7 +1004,7 @@ class TestExtractEntries:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **Tools**
@@ -1031,7 +1031,7 @@ class TestExtractEntries:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            ## Stdlib
@@ -1050,7 +1050,7 @@ class TestExtractEntries:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **Tools**
@@ -81,35 +81,35 @@ MINIMAL_README = textwrap.dedent("""\
    Some intro text.
-    ---
+    ## Projects
-    ## Alpha
+    ### Alpha
    _Libraries for alpha stuff._
    - [lib-a](https://example.com/a) - Does A.
    - [lib-b](https://example.com/b) - Does B.
-    ## Beta
+    ### Beta
    _Tools for beta._
    - [lib-c](https://example.com/c) - Does C.
-    # Resources
+    ## Resources
    Where to discover resources.
-    ## Newsletters
+    ### Newsletters
    - [News One](https://example.com/n1)
    - [News Two](https://example.com/n2)
-    ## Podcasts
+    ### Podcasts
    - [Pod One](https://example.com/p1)
-    # Contributing
+    ## Contributing
    Please contribute!
 """)
@@ -120,11 +120,11 @@ GROUPED_README = textwrap.dedent("""\
    Some intro text.
-    ---
+    ## Projects
    **Group One**
-    ## Alpha
+    ### Alpha
    _Libraries for alpha stuff._
@@ -133,25 +133,25 @@ GROUPED_README = textwrap.dedent("""\
    **Group Two**
-    ## Beta
+    ### Beta
    _Tools for beta._
    - [lib-c](https://example.com/c) - Does C.
-    ## Gamma
+    ### Gamma
    - [lib-d](https://example.com/d) - Does D.
-    # Resources
+    ## Resources
    Where to discover resources.
-    ## Newsletters
+    ### Newsletters
    - [News One](https://example.com/n1)
-    # Contributing
+    ## Contributing
    Please contribute!
 """)
@@ -191,7 +191,7 @@ class TestParseReadmeSections:
            all_names.extend(c["name"] for c in g["categories"])
        assert "Contributing" not in all_names
-    def test_no_separator(self):
+    def test_no_projects_heading(self):
        groups = parse_readme("# Just a heading\n\nSome text.\n")
        assert groups == []
@@ -199,19 +199,19 @@ class TestParseReadmeSections:
        readme = textwrap.dedent("""\
            # Title
-            ---
+            ## Projects
-            ## NullDesc
+            ### NullDesc
            - [item](https://x.com) - Thing.
-            # Resources
+            ## Resources
-            ## Tips
+            ### Tips
            - [tip](https://x.com)
-            # Contributing
+            ## Contributing
            Done.
        """)
@@ -225,15 +225,15 @@ class TestParseReadmeSections:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
-            ## Algos
+            ### Algos
            _Algorithms. Also see [awesome-algos](https://example.com)._
            - [lib](https://x.com) - Lib.
-            # Contributing
+            ## Contributing
            Done.
        """)
@@ -273,17 +273,17 @@ class TestParseGroupedReadme:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **Empty**
            **HasCats**
-            ## Cat
+            ### Cat
            - [x](https://x.com) - X.
-            # Contributing
+            ## Contributing
            Done.
        """)
@@ -295,15 +295,15 @@ class TestParseGroupedReadme:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
            **Note:** This is not a group marker.
-            ## Cat
+            ### Cat
            - [x](https://x.com) - X.
-            # Contributing
+            ## Contributing
            Done.
        """)
@@ -317,19 +317,19 @@ class TestParseGroupedReadme:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
-            ## Orphan
+            ### Orphan
            - [x](https://x.com) - X.
            **A Group**
-            ## Grouped
+            ### Grouped
            - [y](https://x.com) - Y.
-            # Contributing
+            ## Contributing
            Done.
        """)
@@ -405,15 +405,15 @@ class TestParseSectionEntries:
        readme = textwrap.dedent("""\
            # T
-            ---
+            ## Projects
-            ## Async
+            ### Async
            - [asyncio](https://x.com) - Async I/O.
              - [awesome-asyncio](https://y.com)
            - [trio](https://z.com) - Friendly async.
-            # Contributing
+            ## Contributing
            Done.
        """)
@@ -480,21 +480,21 @@ class TestParseRealReadme:
        md = MarkdownIt("commonmark")
        root = SyntaxTreeNode(md.parse(self.readme_text))
-        # Find category section boundaries (between --- and # Resources/Contributing)
+        # Find category section boundaries (between Projects and Resources/Contributing)
-        hr_idx = None
+        projects_idx = None
        end_idx = None
        for i, node in enumerate(root.children):
-            if hr_idx is None and node.type == "hr":
+            if node.type == "heading" and node.tag in ("h1", "h2"):
                hr_idx = i
            elif node.type == "heading" and node.tag == "h1":
                text = render_inline_text(node.children[0].children) if node.children else ""
-                if end_idx is None and text in ("Resources", "Contributing"):
+                if projects_idx is None and text == "Projects":
                    projects_idx = i
                elif end_idx is None and text in ("Resources", "Contributing"):
                    end_idx = i
-        if hr_idx is None:
+        if projects_idx is None:
            return
        bad = []
-        cat_nodes = root.children[hr_idx + 1 : end_idx or len(root.children)]
+        cat_nodes = root.children[projects_idx + 1 : end_idx or len(root.children)]
        for node in cat_nodes:
            if node.type != "bullet_list":
                continue