pypi

Extract PyPI package listings and metadata via RSS feeds and JSON API.

5|Updated Apr 9, 2026

Search Tags:#pypi #scraping #rss #feeds #anti-bot #json-api #package-metadata

Ready to run

npx skills add https://github.com/AgentComputerAI/torch --skill pypi

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: pypi
Source: https://github.com/AgentComputerAI/torch/tree/main/skills/sites/pypi
Command: npx skills add https://github.com/AgentComputerAI/torch --skill pypi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a reliable way to obtain package listings and metadata from pypi.org without being blocked by FullStory client challenges on the /search HTML endpoint, avoiding brittle browser scraping.

Core Features & Use Cases

Feed-first discovery: Use the official RSS feeds (packages.xml, updates.xml, per-project feeds) to reliably discover recent packages and releases.
Rich metadata retrieval: Enrich feed items by calling the JSON API at /pypi/<name>/json to obtain descriptions, authors, classifiers, and release artifacts.
Scaling and backfill: Poll feeds for live updates, use BigQuery for historical data, or iterate the /simple/ index with throttling for larger crawls.
Use Case: Seed a package ingestion pipeline from the RSS feed, dedupe by name, and fetch per-package JSON for downstream indexing.

Quick Start

Use the pypi skill to fetch the latest RSS feed from pypi.org and retrieve detailed package metadata via the JSON API.

Frequently Asked Questions about pypi

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I scrape PyPI package data without getting blocked by anti-bot protections?▼

To scrape PyPI package data without anti-bot blocks, use official RSS feeds like packages.xml and updates.xml for discovery, then fetch rich metadata via the JSON API at pypi.org/pypi/<name>/json with polite throttling.

Why does the PyPI search HTML endpoint return a client challenge instead of results?▼

The PyPI search HTML endpoint returns a client challenge because it is protected by a FullStory anti-bot mechanism, requiring you to use the JSON API or RSS feeds for reliable package discovery instead.

What is the best way to get historical PyPI package metadata for a large backfill?▼

The best way to backfill historical PyPI package metadata is querying BigQuery for bulk historical data, or iterating the /simple/ index with throttling for larger crawls while using a descriptive user-agent.

How do I retrieve detailed release information for a specific Python package from PyPI?▼

To retrieve detailed release information for a specific Python package, request the JSON API endpoint at pypi.org/pypi/<name>/json, which returns descriptions, authors, classifiers, and release artifacts.

Can I use the PyPI RSS feeds to seed a package ingestion pipeline?▼

Yes, you can use the official PyPI RSS feeds to seed a package ingestion pipeline by polling for live updates, deduplicating items by name, and then fetching per-package JSON to enrich the data.

What are the limitations of using the /simple/ index for crawling PyPI packages?▼

Crawling the /simple/ index for PyPI packages requires strict polite throttling and a descriptive user-agent to avoid rate limits, making it slower than RSS feeds or BigQuery for large-scale package discovery.