Lumen
Lumen exposes a self-hosted LMS as an MCP server so Claude can read your course catalog, pull lesson content for RAG-backed tutoring, generate quizzes, and trigger AI authoring workflows. The demo runs on a single t4g.small with Groq Llama 3.3 70B, but the provider layer swaps to Anthropic or OpenAI via environment config. Under the hood it's FastAPI + Celery + Postgres with pgvector, Redis Streams for SSE, and a custom orchestrator that routes between five sub-agents (retriever, web searcher, code runner, quiz gen, concept explainer). Every LLM call hits a cost meter and lands in an observable trace table. You'd reach for this if you're building educational tooling and want Claude to interact with lesson content, spaced-repetition queues, or multi-agent tutoring sessions without writing your own RAG plumbing.
- GitHub stars
- 69
- tools
- 9
- last updated
- Jun 7, 2026
Tools
9 toolsPublic tool metadata for what this MCP can expose to an agent.
list_coursesCatalog query
Catalog query
No parameter schema in public metadata yet.
get_courseCourse + syllabus tree
Course + syllabus tree
No parameter schema in public metadata yet.
search_lesson_contentSemantic search over a course's lesson chunks
Semantic search over a course's lesson chunks
No parameter schema in public metadata yet.
ask_tutorCourse-scoped RAG tutor with cited answer
Course-scoped RAG tutor with cited answer
No parameter schema in public metadata yet.
list_my_due_reviewsFSRS-6 spaced-repetition queue
FSRS-6 spaced-repetition queue
No parameter schema in public metadata yet.
grade_review_cardSubmit one rating (1=again, 2=hard, 3=good, 4=easy)
Submit one rating (1=again, 2=hard, 3=good, 4=easy)
No parameter schema in public metadata yet.
list_my_progressPer-course completion + mastery rollup
Per-course completion + mastery rollup
No parameter schema in public metadata yet.
create_course_draftAI-authored outline + draft course
AI-authored outline + draft course
No parameter schema in public metadata yet.
ingest_url_to_draftMulti-modal ingest (YouTube / Notion / Google Docs)
Multi-modal ingest (YouTube / Notion / Google Docs)
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
Lumen
Type a one-sentence learning goal. An AI orchestrator builds you a private course (45–90 s for 10 lessons), a course-scoped RAG tutor teaches it with lesson citations, and its agent steps and LLM calls are logged.
Custom multi-agent system, no LangChain · public evals with the weak scores kept in · ran in production on AWS, May to October 2026
-2ea44f?style=flat-square){kind=icon}
Details
- Repository
- github.com/ahmedeid1/lumen
- Website
- github.com/ahmedEid1/lumen/blob/main/docs/mcp.md
- Category
- AI & LLM Tools
- Registry
- MCP registry, active
More from ahmedeid1
Related AI & LLM Tools MCP servers
View all
DOMShellapireno/domshellDrive Chrome with filesystem commands (ls, cd, grep, click, type). One MCP tool, multi-agent.59stars
MCP AutoMemverygoodplugins/mcp-automemGraph-vector memory for AI assistants using FalkorDB and Qdrant51stars
Engramtstockham96/engramIntelligent agent memory with automatic extraction, consolidation, and bi-temporal recall.39stars
Local FAISS MCP Servernonatofabio/local_faiss_mcpLocal FAISS vector database for RAG with document ingestion, semantic search, and MCP prompts.30stars
SAME - Stateless Agent Memory Enginesgx-labs/statelessagentPersistent memory for AI coding agents. Local-first SQLite + vector search. 17 MCP tools.19stars
Promptheusabhichandra21/promptheusAI-powered prompt refinement with adaptive questioning and multi-provider support (6+ LLM backends)17stars