MCP 服务器

Coursera MCP

Enables AI agents to programmatically interact with Coursera, including enrolling in courses, completing lectures, solving quizzes, submitting assignments, earning certificates, and pushing them to LinkedIn.

README

The most advanced MCP server for Coursera. 87 tools · 8 integration layers · full course automation · Obsidian export

 ██████╗ ██████╗ ██╗   ██╗██████╗ ███████╗███████╗██████╗  █████╗       ███╗   ███╗ ██████╗██████╗
██╔════╝██╔═══██╗██║   ██║██╔══██╗██╔════╝██╔════╝██╔══██╗██╔══██╗      ████╗ ████║██╔════╝██╔══██╗
██║     ██║   ██║██║   ██║██████╔╝███████╗█████╗  ██████╔╝███████║█████╗██╔████╔██║██║     ██████╔╝
██║     ██║   ██║██║   ██║██╔══██╗╚════██║██╔══╝  ██╔══██╗██╔══██║╚════╝██║╚██╔╝██║██║     ██╔═══╝
╚██████╗╚██████╔╝╚██████╔╝██║  ██║███████║███████╗██║  ██║██║  ██║      ██║ ╚═╝ ██║╚██████╗██║
 ╚═════╝ ╚═════╝  ╚═════╝ ╚═╝  ╚═╝╚══════╝╚══════╝╚═╝  ╚═╝╚═╝  ╚═╝      ╚═╝     ╚═╝ ╚═════╝╚═╝

Give any AI agent — Claude, GPT, or your own — full programmatic access to Coursera. Enroll in courses, speedrun lectures, solve quizzes, submit assignments, earn certificates, and push them to LinkedIn. All from a single MCP server.

Highlights

	Feature	How
🚀	Speedrun entire courses	`speedrun_course` completes lectures, quizzes, discussions, and graded exams in one call
🔓	Freemium bypass	Auto-detects `enrolledViaFreemiumAt` restrictions and re-enrolls without the flag
🧠	Solve quizzes via GraphQL	Elimination-learning solver: submits, learns from feedback, retries until passing
📝	Submit everything	Written, peer review, discussion prompts, programming assignments — all automated
📊	Full analytics	Learner stats, XP, streaks, goals, module progress, detailed grades, timelines
🔗	LinkedIn integration	Auto-add certificates to your profile with duplicate detection
🕷️	Scrapling scraping	Adaptive parser (1775× faster than BeautifulSoup) + JS rendering
🛡️	Stealth browser	rebrowser-puppeteer + fingerprint injection defeats Cloudflare/bot detection
🔑	Cookie-based auth	No passwords stored — session cookies via Cookie-Editor
📓	Obsidian export	Summarize courses into structured study notes and write directly to your Obsidian vault

Prerequisites

1. Install Cookie-Editor

All authentication uses browser session cookies. Install Cookie-Editor to export them:

Browser	Install
<img src="https://img.shields.io/badge/Chrome-4285F4?style=flat-square&logo=googlechrome&logoColor=white">	Chrome Web Store
<img src="https://img.shields.io/badge/Firefox-FF7139?style=flat-square&logo=firefoxbrowser&logoColor=white">	Firefox Add-ons
<img src="https://img.shields.io/badge/Edge-0078D7?style=flat-square&logo=microsoftedge&logoColor=white">	Edge Add-ons

2. Export Coursera Cookies

Log into coursera.org
Click Cookie-Editor icon → Export → copy CAUTH and CSRF3-Token values
Paste into .env (see Setup)

3. Export LinkedIn Cookies (optional)

Log into linkedin.com
Click Cookie-Editor icon → Export → copy li_at and JSESSIONID values
Paste into .env

Setup

git clone https://github.com/VoidChecksum/coursera-mcp.git
cd coursera-mcp

uv venv .venv
uv pip install -p .venv/bin/python mcp curl_cffi skipera requests "scrapling[fetchers]"
.venv/bin/scrapling install

npm install rebrowser-puppeteer puppeteer-extra puppeteer-extra-plugin-stealth \
            fingerprint-generator fingerprint-injector

cp .env.example .env
chmod 600 .env

`.env` format

COURSERA_CAUTH=<paste CAUTH cookie value>
COURSERA_CSRF3=<paste CSRF3-Token cookie value>

LINKEDIN_LI_AT=<paste li_at cookie value>
LINKEDIN_JSESSIONID=<paste JSESSIONID value, without quotes>

[!NOTE] When Coursera or LinkedIn logs you out, re-export cookies with Cookie-Editor and update .env.

Connect to Your AI Client

<details> <summary>Claude Desktop / Claude Code</summary>

Add to your MCP config (claude_desktop_config.json or via CLI):

{
  "mcpServers": {
    "coursera": {
      "command": "/path/to/coursera-mcp/run.sh"
    }
  }
}

Or via CLI:

claude mcp add coursera /path/to/coursera-mcp/run.sh

</details>

<details> <summary>Other MCP Clients</summary>

Any MCP-compatible client that supports stdio transport works. Point it at run.sh which loads .env and starts the server.

</details>

Tools (87)

Course Discovery & Enrollment

Tool	Description
`whoami`	Verify session, return user ID
`my_courses`	List enrolled courses (name, slug, role)
`search`	Search Coursera catalog via GraphQL
`course_details`	Full course metadata by slug
`enroll_course`	Enroll in a course, auto-bypassing freemium restrictions
`unenroll_course`	Delete course enrollment
`batch_enroll`	Enroll in multiple courses at once with freemium bypass
`reset_course_deadlines`	Reset/extend deadlines to unlock expired submissions

Speedrunning

Tool	Description
`speedrun_course`	Complete an entire course in one call: freemium bypass → lectures → quizzes → discussions → graded exams with elimination learning
`speedrun_all`	Speedrun every enrolled course, skipping 100% courses
`complete_course_lectures`	Mark all lectures, readings, widgets, and labs as completed in parallel

Progress, Grades & Certificates

Tool	Description
`get_course_grades`	Grades, outcomes, and passing state for all graded items
`get_course_progress`	Overall progress % and per-item completion status
`get_detailed_grades`	Full grade breakdown with item outcomes and assignment group grades
`get_module_progress`	Per-module completion status (completed/in-progress/not-started)
`get_home_progress`	Modules completed and modules passed counts
`get_certificates`	All certificates with codes and verification URLs
`get_certificate_details`	Certificate grade, grant date, verification URL, preview image
`get_certificate_image`	Certificate preview image URL

Learner Analytics (Reverse-Engineered GraphQL)

Tool	Description
`get_learner_stats`	Total XP, items completed, learning minutes, active days, weekly stats
`get_weekly_streaks`	Current and target weekly learning streaks
`get_weekly_goals`	Day-by-day goal tracking and progress status
`get_course_timeline`	Start/end dates, ahead/behind/overdue schedule status
`get_profile`	User ID, full name, locale, timezone, profile photos
`get_subscription_status`	Coursera Plus/trial/payment status and subscription details
`check_product_ownership`	Enrollment type (freemium/paid/Plus), ownership status
`send_learning_heartbeat`	Send learning time heartbeat (registers active time on a course item)
`get_cohorts`	Social learning cohorts the user belongs to
`get_item_deadlines`	Per-item content deadlines (more granular than session deadlines)

Skill & Career Intelligence

Tool	Description
`get_skill_xp`	Skill area XP breakdown per course (gained vs potential)
`get_item_xp`	XP earned vs potential for a specific course item
`get_learning_plan`	AI-generated personalized learning plan (role, skills, courses)
`get_related_courses`	Related course recommendations
`get_interested_careers`	User's interested career roles
`get_career_roles`	Career role recommendations for a specialization
`get_specialization_progress`	Progress across all courses in a specialization
`get_course_info`	Course description, partners, instructors, workload, primary specialization

Quiz & Assessment (GraphQL — no browser)

Tool	Description
`get_quiz_questions_api`	Fetch questions + option IDs via GraphQL, handles START_NEW_ATTEMPT
`submit_quiz_answers_api`	Submit answers, get instant grade + per-question feedback
`get_quiz_state`	Check quiz attempt status, grade, and rate-limit info without starting
`get_discussion_prompt`	Get discussion prompt content and metadata
`submit_discussion_prompt_answer`	Post answer to discussion prompt

Content Extraction

Tool	Description
`get_video_transcript`	Clean text transcript of any lecture video (VTT dedup)
`get_supplement_content`	Clean text of any reading/supplement
`get_lecture_assets`	Downloadable slides, PDFs, and supplementary files
`get_course_references`	Glossary and reference materials
`get_instructor_notes`	Module-level instructor notes
`get_learning_objectives`	Module learning objectives
`get_user_notes`	Personal study notes for a course item
`create_user_note`	Create study notes on lectures/readings with video timestamps

Course Structure & Navigation

Tool	Description
`course_materials`	Full course syllabus: modules, lessons, items
`module_items`	Items in a specific module (lectures, quizzes, readings)
`get_course_navigation`	Full navigation with deadline status per module
`get_course_schedule`	Session deadlines per module and item
`get_forum_structure`	Full forum hierarchy (IDs, types, descriptions)
`get_specialization_info`	Specialization courses, partners, pricing
`get_career_roles`	Career role recommendations for specializations
`get_ai_grading_eligibility`	Check if AI peer review grading is available

Browser Automation (stealth Chromium)

Tool	Description
`get_quiz_attempt`	Start/resume quiz in browser, extract all questions
`submit_quiz_attempt`	Answer quiz, sign honor code, submit
`get_assignment_instructions`	Extract assignment instructions + form fields
`submit_assignment`	Submit written/peer assignment (text + files)
`get_workspace_url`	Get Jupyter/lab workspace URL
`mark_item_completed`	Trigger completion by loading item page
`get_forum_threads`	Extract discussion forum threads
`post_forum_thread`	Post new thread in forums
`get_peer_review_questions`	Get peer review criteria + rubrics
`submit_peer_review_evaluation`	Submit peer review grading
`get_programming_assignment_instructions`	Get programming assignment schema
`submit_programming_api`	Submit code to Coursera autograder

Scrapling (adaptive web scraping)

Tool	Description
`scrapling_api_fetch`	Fast API call via Scrapling (stealth httpx)
`scrape_course_page`	Scrape authenticated pages with JS rendering (Playwright)
`scrape_url`	General-purpose scraper with CSS selectors

LinkedIn Integration

Tool	Description
`get_linkedin_certifications`	Fetch existing certs on LinkedIn profile
`add_certificate_to_linkedin`	Add cert to LinkedIn (auto-skips duplicates)
`add_all_certificates_to_linkedin`	Batch-add all Coursera certs (dedup)

Study & Planning

Tool	Description
`course_content_digest`	Bundle a module's transcripts + readings into one study document
`search_course_transcripts`	Full-text search across a course's lecture transcripts
`learning_dashboard`	Progress %, certificate, and grade across all enrolled courses
`get_upcoming_deadlines`	Session end dates across enrolled courses, soonest first
`export_deadlines_ics`	Export deadlines as an importable `.ics` calendar
`get_specialization`	Specialization details + its course list

Obsidian Integration

Tool	Description
`summarize_module`	Summarize a module's lectures and readings into concise study notes
`summarize_course`	Summarize an entire course organized by module
`export_course_to_obsidian`	Summarize + write to Obsidian vault with frontmatter, wiki-links, callouts, and tags
`export_module_to_obsidian`	Export a single module to Obsidian
`export_all_courses_to_obsidian`	Batch export all enrolled courses to Obsidian

Reverse Engineering

Several tools were built by intercepting live browser traffic and reverse-engineering Coursera's internal APIs:

Freemium Bypass — Coursera sets enrolledViaFreemiumAt on free enrollments, which causes all opencourse.v1 video-event and supplement-completion endpoints to return 403. The bypass deletes the enrollment via DELETE /api/openCourseMemberships.v1/{uid}~{courseId} (with an empty {} body — Akka backend requires it) and re-creates it via POST without the freemium flag.

GraphQL Gateway — 48+ operations captured from graphql-gateway, including GetLearnerStats, WeeklyStreaks_QueryWeeklyStreaks, GetLearnerWeeklyGoals, GetCourseBranchModuleProgressByCourse, GetLearnerCourseTimeline, GetMyCertificateByProduct, CourseNavigation, GetAiGradingEligibilityByCourse, CareerRolesQuery, and GetProductOwnership.

REST APIs — 61+ endpoints mapped from browser network traffic, including onDemandCourseViewGrades.v1, onDemandReferences.v1, onDemandLectureAssets.v1, onDemandInstructorNotes.v1, userNotes.v1, onDemandCourseForums.v1, onDemandLearningObjectives.v2, and subscriptions.v1.

Quiz Solver — Elimination-learning algorithm: submits first-option answers, gets per-question feedback marking options correct/incorrect, persists learned answers to ~/.skipera/gradedData/, and retries with improved answers until passing (≥70%). Handles START_NEW_ATTEMPT, rate limits, and MultipleChoiceReflectQuestion types.

Architecture

coursera-mcp/
├── server.py            Python MCP server — all 87 tools
├── automation.mjs       Stealth browser automation (Node.js)
├── run.sh               Launcher (loads .env, starts stdio)
├── .env                 Your cookies (gitignored)
├── .env.example         Template
└── .gitignore

┌──────────────────────────────────────────────────────────────────────┐
│              MCP Clients (Claude / GPT / Custom)                     │
├──────────────────────────────────────────────────────────────────────┤
│                server.py  ·  FastMCP stdio                           │
├─────────┬─────────┬──────────┬───────────┬──────────┬────────┬───────┤
│  REST   │ GraphQL │ Scrapling│  Browser  │ LinkedIn │ gRPC   │Obsidi-│
│  API    │ Gateway │          │  Stealth  │  Voyager │  API   │an API │
│         │         │          │ Puppeteer │ + Browser│        │       │
├─────────┴─────────┤          ├───────────┤          │        │       │
│  curl_cffi        │ Fetcher  │automation │automation│        │ REST  │
│  skipera          │ Dynamic  │   .mjs    │   .mjs   │        │:27123 │
│  requests         │ Fetcher  │ (Node.js) │ (Node.js)│        │       │
└───────────────────┴──────────┴───────────┴──────────┴────────┴───────┘

Security


No passwords stored	Cookie-based auth only — session tokens from your browser
`.env` is gitignored	Secrets never enter version control
No hardcoded credentials	All secrets read from `os.environ` at runtime
No secret logging	Cookies are never printed or logged to stdout/stderr
`.env.example` is safe	Contains only empty placeholders

[!CAUTION] Never commit your .env file. If you accidentally expose cookies, log out of Coursera/LinkedIn immediately to invalidate them.

License

MIT

Built for agents that learn faster than humans.