LinkedIn Learning

🎓 LinkedIn Learning API: Course Data, Reviews and Syllabus from Python and MCP

A Python and MCP quick-start for the LinkedIn Learning API on Apify. Search the public LinkedIn Learning course catalog by keyword, skill level, length or software, then pull the full record for any course: rating and rating count, written learner reviews, the complete syllabus with a description and a free-preview flag on every lesson, instructors with profile links, skills taught and certificate details. None of it needs a login, a cookie or a site license.

• Actor: LinkedIn Learning API on Apify
• Input schema: input parameters
• Get a free API token: apify.com

LinkedIn's own Learning API is available only through its Partner Program or a purchased site license, with OAuth keys an admin has to provision. This Actor reads the pages LinkedIn Learning already publishes to logged-out visitors and returns the same catalog metadata as JSON: course title, link, instructors, duration, viewer count and thumbnail from a search, or the full record with reviews and table of contents from a course URL. It covers the public catalog only. LinkedIn's reporting API, which returns learner activity for a licensed organization, is a different product and nothing here reads it. You pay per row: $0.10 per 1,000 search results and $0.50 per 1,000 full course records. Apify also applies its platform accounting events, $0.00001 per Actor start (one event per GB of run memory, at least one) and $0.00001 per stored dataset row.

Video: Build MCP servers with Apify

Text walkthrough

The LinkedIn Learning API has two modes, set by mode. In search mode you pass queries (up to 25 keywords) or topics, optionally narrowed by sortBy, difficultyLevel, entityType, duration and softwareNames, and each course comes back as a search_result row with title, courseUrl, courseId, entityType, instructors, durationText, viewersText, releaseText and thumbnailUrl. LinkedIn serves at most 50 results per query, so expandWithFilters re-runs the query across LinkedIn's own filter combinations and merges the unique courses when you need a fuller course list. In details mode you pass courseUrls (course, lesson or learning path links, up to 200) and get a course_detail row with ratingValue, ratingCount, enrollmentCount, reviews, tableOfContents with a per-lesson isFree flag, skills, instructors, lessonCount, freeLessonCount, hasCertificate and certificateName. Setting enrichDetails to true on a search opens each found course and returns the full record instead, billed once at the full-record rate. The published task Extract LinkedIn Learning Course Reviews and Ratings runs details mode on three course URLs and returns the written review text with each reviewer's name, job title and profile link; run_course_reviews() in the example file makes the same details-mode call from Python, trimmed to the first of those three URLs so the first run costs about $0.0005.

Quick Start

Prerequisites: Python 3.11 or newer, uv, and a free Apify API token from apify.com.

git clone https://github.com/johnisanerd/Apify-LinkedIn-Learning-API.git

cd Apify-LinkedIn-Learning-API uv sync cp .env.example .env # paste your token into .env uv run python linkedin-learning-api-example.py

Each example is a separate flag:

uv run python linkedin-learning-api-example.py --example default      # search "python", 5 rows

uv run python linkedin-learning-api-example.py --example catalog # course catalog rows, each printed as JSON uv run python linkedin-learning-api-example.py --example reviews # written reviews for one course uv run python linkedin-learning-api-example.py --example new-courses # newest releases on a topic uv run python linkedin-learning-api-example.py --example all

Every example asks for a small number of rows on purpose. You pay per row delivered, so confirm the shape of the data first, then raise maxItems.

If you do not want a .env file, export the token instead:

export APIFY_API_TOKEN="paste_your_real_token_here"   # replace with the token from your Apify account

uv run python linkedin-learning-api-example.py

Why use this LinkedIn Learning API

A short list of what the Actor returns and what it costs, before the field-level detail in Features and Output Format.

No login and no site license

The Actor reads only what LinkedIn Learning shows a logged-out visitor. There is no field for a session cookie and no OAuth step. Anyone with a free Apify token can run it.

The full syllabus, lesson by lesson

tableOfContents lists every section and lesson with its own title, description, durationSeconds, url and an isFree flag, with freeLessonCount totalled on the course. That is how you find the free preview lessons without opening each page.

Pay for the rows you receive

Search results are billed on the course-found event at $0.10 per 1,000; full records on course-detail at $0.50 per 1,000. Duplicates, error rows and the popular-courses page LinkedIn substitutes when a query matches nothing are never charged at the course-found or course-detail rate.

LinkedIn Learning reviews with the reviewer's identity

A full course record carries reviews whenever the course page publishes written reviews: the review text, its star rating, and the reviewer's authorName, authorJobTitle and authorProfileUrl, plus datePublished. Other course data tools stop at the star average and rating count; the written text and who wrote it are only on the course page.

LinkedIn Learning course catalog and course list past the 50-result cap

LinkedIn shows at most 50 results for any one search. Run search mode across a set of keywords or topic pages, turn on expandWithFilters to get past that ceiling, and export the dataset as JSON, CSV or Excel. expandWithFilters repeats each query across LinkedIn's own filter combinations and merges the unique courses; within each query rows deduplicate on courseId, lessons on their own URL, so a course reached through two slugs is returned once; a course that matches two different queries appears once under each searchQuery. It is the closest thing to a LinkedIn Learning course catalog download that needs no account.

LinkedIn Learning learning paths, lessons and courses, told apart

entityType separates COURSE, VIDEO and LEARNING PATH in search results. A learning path URL in details mode returns pathUrl, courseCount and the ordered courses inside it; a lesson URL returns that lesson joined to its parent courseId.

Features

Core Capabilities

• Search by queries or topics, narrowed with sortBy, difficultyLevel, entityType, duration and softwareNames.
• LinkedIn Learning course catalog download: expandWithFilters re-runs a query across LinkedIn's own filter combinations, the only route past the 50-results-per-query cap, so a course list longer than 50 rows comes from one run.
• LinkedIn Learning learning paths: details mode takes course, lesson and learning path URLs, up to 200 per run, and returns a course_detail, lesson or learning_path row for each.
• enrichDetails turns a search into full course records in one run, billed once per row at the full-record rate.
• LinkedIn Learning free courses: freeLessonCount and the per-lesson isFree flag in tableOfContents show how much of each course is watchable without a subscription.
• Online courses dataset: run many queries or topics in one search and export the rows from every query, each labelled with its searchQuery, as JSON, CSV or Excel.

Data Quality

• Each review carries body, rating, authorName, authorJobTitle, authorProfileUrl and datePublished.
• Every lesson in tableOfContents has its own description, durationSeconds, url and isFree flag.
• courseUrl is the canonical link with tracking parameters stripped; within each query (and within a details run) rows deduplicate on courseId, lessons on their own URL, so a course reached through two slugs is returned once; a course that matches two different queries appears once under each searchQuery.
• no_results and error rows are pushed so you can see what failed, and are never billed at the course-found or course-detail rate.

Recipes

Ready-made configurations with their own Store landing pages:

• LinkedIn Learning Course Data as JSON, No Login: search mode on "project management" with maxItems 25; each row has title, link, instructors, duration, viewer count and thumbnail. Local: uv run python linkedin-learning-api-example.py --example catalog
• LinkedIn Learning Course Data Without a Login: the same search shape on "cybersecurity", proving the catalog reads with no account and no session cookie.
• Extract LinkedIn Learning Course Reviews and Ratings: details mode on three course URLs; each row carries ratingValue, ratingCount and the written reviews. Local: uv run python linkedin-learning-api-example.py --example reviews
• Build a LinkedIn Learning Course Ratings Dataset: search "python" with enrichDetails true and maxItems 5, so each course row is a full record with rating, rating count, learner numbers, level and duration (learning path hits stay as search rows).
• Track New LinkedIn Learning Course Releases: search "artificial intelligence" with sortBy RECENCY, newest first. Local: uv run python linkedin-learning-api-example.py --example new-courses

Schedule tip. Save any of these inputs as a Task on the Actor page and schedule it to run weekly. With sortBy set to RECENCY the dataset lists the newest courses on a topic every week, and with enrichDetails on it refreshes ratingValue, ratingCount and the written reviews for every course, without anyone touching it.

Usage Examples

Basic, matching the default run:

{

"mode": "search", "queries": ["python"], "sortBy": "RELEVANCE", "difficultyLevel": "ANY", "entityType": "COURSE", "duration": "ANY", "maxItems": 5 }

Advanced, a filtered search enriched to full records:

{

"mode": "search", "queries": ["excel"], "softwareNames": ["Microsoft Excel"], "difficultyLevel": "BEGINNER", "duration": "BETWEEN_1_TO_2_HOURS", "sortBy": "POPULARITY", "maxItems": 10, "enrichDetails": true, "maxConcurrency": 5 }

Details mode, full records for URLs you already have:

{

"mode": "details", "courseUrls": [ "https://www.linkedin.com/learning/python-essential-training-18764650", "https://www.linkedin.com/learning/paths/getting-started-with-python" ] }

Input Parameters

Only mode is required. Search settings apply when mode is search; courseUrls applies when mode is details. In search mode at least one of `qu