PyPI Package Tracker — Metadata + Stats

Snapshot PyPI package metadata using the public pypi.org/pypi/<pkg>/json endpoint. One row per package: version, classifiers, requires_python, maintainers, project URLs, and release count. No authentication required.

---

Example output row

{
  "snapshotAt": "2026-06-03T17:10:00+00:00",
  "name": "requests",
  "latestVersion": "2.32.3",
  "summary": "Python HTTP for Humans.",
  "author": "Kenneth Reitz",
  "authorEmail": "me@kennethreitz.org",
  "license": "Apache 2.0",
  "homepage": "https://requests.readthedocs.io",
  "projectUrls": {
    "Documentation": "https://requests.readthedocs.io",
    "Source": "https://github.com/psf/requests"
  },
  "requiresPython": ">=3.8",
  "classifiers": [
    "License :: OSI Approved :: Apache Software License",
    "Programming Language :: Python :: 3"
  ],
  "keywords": "http client requests",
  "releaseCount": 142
}

---

How to use

Input schema

Field	Type	Default	Description
packages	string[]	["requests","numpy","pandas"]	PyPI package names to snapshot

Minimal input example

{
  "packages": ["httpx", "pydantic", "fastapi"]
}

Pass any number of package names. Packages that return 404 (deleted or misspelled) are silently skipped — they do not count as a billed row.

---

Pricing

Event	Description	Price
package_snapshot	One PyPI package metadata row.	$0.0002

You are charged only for rows successfully fetched and pushed to the dataset. A 20-package run costs ~$0.004.

---

Buyer

• Python developers tracking dependency metadata across a project's requirements tree.
• Security teams auditing requires_python constraints and classifier flags across an internal package inventory.
• OSS analysts monitoring version cadence and release counts for ecosystem health reports.
• Data engineers building pip-audit pipelines or license-compliance dashboards.
• Researchers collecting classifier and keyword distributions across the PyPI corpus.

---

Source

Data is fetched from the official PyPI JSON API:

GET https://pypi.org/pypi/{package}/json

No authentication. Rate-limited to one request every 250–600 ms (polite crawler policy). PyPI's terms of service permit programmatic metadata access for non-abusive use.