Wikify

Compile raw sources into detailed, Obsidian-compatible, interlinked Markdown pages under wiki/ references/ and concepts/.

Deduplicates concepts, splits overly broad concepts, and synthesizes multi-source concept definitions by dynamically searching and analyzing all papers that reference them.

Ingest new academic papers or PDFs into the raw/ folder of your active topic wiki using the local OCR model configured in config.yaml.

Ingest new academic papers, notes, or web articles into the raw/ folder of your active topic wiki.

Initialize a new topic workspace folder with standard raw/, wiki/, inbox/, and output/ directories.

Statically check and repair double-bracket linkages and frontmatter within your compiled wiki directory.

Automatically builds semantic links between concept markdown files by calculating vector similarity using a local Ollama embedding model.

Act as a Concept Miner to inspect compiled papers, check concept linkage density, and spawn subagents to extract missing mathematical and physical concepts.

Chat with the knowledge base. Answers user questions by dynamically employing RAG context extraction, graph database queries, and targeted keyword searches to provide cited, hallucination-free answers.

Audit the compiled wiki pages to cross-check statements, highlight scientific contradictions, and output theses using a Map-Reduce architecture.

Extract Obsidian-style relationships (wikilinks, tags, aliases) into a structured AI-friendly SQLite graph database.

Initialize a central Hub directory for managing multiple topic wikis.

Manage topics in the central Hub: route slugs to paths, list topics, and archive or restore topics.

Spawn parallel academic subagents to perform multi-perspective research on a given query and compile a detailed synthesis report.

Deduplicates, normalizes, and reduces sprawling tags and aliases across the knowledge graph using a Map-Reduce architecture.