skills.rest
jorgearango
jorgearangoCOMMUNITY
Explore all skills by this author →
Website𝕏@jarangoGitHub

llmapper

Generate concept maps from articles via a multi-stage LLM pipeline producing RDF and Mermaid diagrams.

16|Updated Nov 5, 2025
Search Tags:#knowledge graph#mermaid#visualization#concept map#rdf#article summarization#llm pipeline
One-click install
npx skills add https://github.com/jorgearango/llmapper-skill --skill llmapper-jorgearango
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: llmapper
Source: https://github.com/jorgearango/llmapper-skill
Command: npx skills add https://github.com/jorgearango/llmapper-skill --skill llmapper-jorgearango

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning long articles into clear visual knowledge graphs is time-consuming and subjective. This Skill automates the process by transforming articles into concept maps that answer what a subject is about and why it matters, using a structured four-stage pipeline. ## Core Features & Use Cases - Multi-Format Input: Accepts file uploads (PDFs, text files, Word documents), URLs with full-text validation, and pasted text. - Four-Stage Pipeline: Generates three focusing questions for user selection, extracts concepts via a panel-of-experts summarization, builds an RDF knowledge graph as the canonical source of truth, and renders a Mermaid flowchart inline with a saved .mermaid file. - Iterative Refinement: All user-requested changes are applied to the stored RDF first, then the visualization is regenerated, ensuring consistency across edits. - Use Case: Upload a research paper PDF, pick one of three focusing questions, and receive a styled Mermaid concept map rendered inline plus saved to /tmp for reuse, with an optional interactive Cytoscape HTML alternative. ## Quick Start Ask Claude to create a concept map from an uploaded article file, a URL, or pasted text, then choose one of the three focusing questions presented.

Frequently Asked Questions about llmapper

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a concept map from an article?▼

Provide the article as a file upload, URL, or pasted text and ask for a concept map. The skill generates three focusing questions, you pick one, and it produces a Mermaid diagram rendered inline plus a saved .mermaid file in /tmp.

What input formats does the concept map skill support?▼

It accepts PDF, text, and Word file uploads as the primary method, URLs fetched via WebFetch with summary detection, and pasted text as a fallback. File upload is the most reliable since some sites block web fetching.

Can I edit a concept map after it is generated?▼

Yes. Request changes in natural language, such as adding a node or removing a relationship. The skill modifies the stored RDF knowledge graph first, then regenerates the visualization from the updated RDF.

Why does the same article produce different concept maps?▼

The pipeline is non-deterministic by design. The panel-of-experts summarization explores different perspectives on why a subject matters, so each run can yield different valid maps, especially with different focusing questions.

What are the limitations of Mermaid concept map rendering?▼

Mermaid automatic layout can become cluttered with very large graphs of 30 or more nodes. For complex graphs, the skill offers an alternative Cytoscape.js HTML output with interactive drag, zoom, and pan saved to /tmp.

Related Skills

Explore More Education & Research Skills→
vibewatchvibewatch

fetch-url

Fetch URLs and extract readable text from HTML pages and PDFs.

Official
Intermediate
xjtulycxjtulyc
33

knowledge-graph-sparql

Build RDF knowledge graphs from research metadata and query them with SPARQL.

Community
Advanced
retranretran
34

resources

Extract and link concepts from Confluence pages and Jira issues into a knowledge graph.

Community
Advanced
Yoodaddy0311Yoodaddy0311
3

oss-ai-catalog

Recommends open-source AI tools, models, and frameworks from a curated category index.

Community
Basic