How do their APIs compare?

I can only answer questions about practical-tutorials/project-based-learning or vinta/awesome-python. Neither of these repositories exposes an "API" in the sense of a service or library API for consumers — let me clarify what's actually in each repo based on the retrieved data.

practical-tutorials/project-based-learning: This is a curated list (README.md) of tutorial links, not a software library. It has no runtime API. The only "code" present is internal tooling — scripts/check_readme.py, which validates the README's structure and link health. It contains helper functions like:

• classify_line(line) — parses a single README line into a tuple classified as 'blank' | 'header' | 'entry' | 'series' | 'toc' | 'unparseable' [scripts/check_readme.py:L101-L135]
• normalize_no_scheme(url) — strips the scheme from a URL for comparison purposes [scripts/check_readme.py:L187-L191]
• _row(state, url) — looks up a URL's tracked state (used for link-rot tracking) [scripts/check_readme.py:L882-L886]
• load_url_items_from_file(path) — loads URL items either from a plain list or a check-diff --json-style dict with added_urls [scripts/check_readme.py:L772-L789]

These are internal maintenance scripts (used by the link-rot CI workflow referenced in the README badge) rather than a public API meant for external consumption [README.md excerpt].

vinta/awesome-python: Similarly, this is primarily a curated README list of Python libraries (e.g., the Web APIs section listing frameworks like fastapi, django-rest-framework, falcon, etc. [README.md:L266-L284]). But this repo also has a small website/tooling layer:

• website/static/main.js — client-side JS powering the site's table filtering, sorting, and search (functions like applyFilters() [website/static/main.js:L133-L192], sortRows() [website/static/main.js:L248-L283], and tag/filter click handlers [website/static/main.js:L336-L352]). This isn't a backend API either — it's DOM manipulation logic for the static site frontend.
• website/tests/test_fetch_github_stars.py — tests for a parse_graphql_response function that parses GitHub's GraphQL API responses to extract star counts, owner logins, and last-commit timestamps for listed repos [website/tests/test_fetch_github_stars.py:L105-L159]. This is the closest thing to "API" code in either repo — it's a consumer of GitHub's GraphQL API, not a served API itself.

Comparison summary: Neither repo ships a public API. project-based-learning has minimal internal tooling for README/link validation, while awesome-python has a bit more supporting infrastructure — a static website with client-side JS for interactivity and a script that consumes GitHub's GraphQL API to enrich listings with star counts and activity data.

FOLLOW_UPS:

• How does the link-rot checker classify a broken URL?
• What does parse_graphql_response do with missing owner data?
• How does main.js implement search filtering on the site?