Generate a knowledge graph from a directory of files and explore relationships.
npx skills add https://github.com/lyingparachute/skills --skill graphify-lyingparachute
Please help me install this Agent Skill. Skill: graphify Source: https://github.com/lyingparachute/skills/tree/main/graphify Command: npx skills add https://github.com/lyingparachute/skills --skill graphify-lyingparachute
This Skill enables users to quickly and easily create a knowledge graph from any collection of files, including code, documents, papers, images, and videos. It helps users understand the relationships between different elements and answer questions about the codebase or project.
To create a knowledge graph from a directory of files, use the graphify skill with the path to the directory: /graphify path/to/files.
High-intent search queries and answers about installing and using this skill.
To generate a knowledge graph from a directory, use the graphify skill with the target path. It automatically parses code, documents, papers, images, and videos to build a persistent graph of relationships.
Community detection in codebase analysis identifies and labels clusters of related concepts within a knowledge graph. This helps you understand structural relationships and navigate complex project dependencies more effectively.
You can analyze images and videos alongside code and documents. The knowledge graph creation supports various file types by parsing multimedia content using libraries like OpenCV and TensorFlow.
Graph traversal and analysis require the graphify library along with networkx, pandas, numpy, and scikit-learn. These dependencies enable interactive exploration and querying of the generated graph.
The best way to export a knowledge graph for external visualization is using the built-in export feature. It supports outputting the graph structure in JSON, SVG, and GraphML formats.
Graph analysis works with unstructured PDF and Word documents by extracting text and metadata. It uses pdfplumber and python-docx to parse these files into nodes within the knowledge graph.
lousy-agentsAnalyze and visualize codebase dependencies using NetworkX directed graphs.
Abulfadl-AhmadiConvert an Obsidian wiki knowledge graph into JSON, GraphML, Cypher, and HTML.
Ritabrata-ChakrabortyAutomate machine learning workflows with scikit-learn preprocessing, model selection, and evaluation.
bmaddevelopment-stackCombines vector search with graph traversal for analyzing codebases.
