mj-deving /graphify
Turn any folder of files into a navigable knowledge graph with community detection, an honest audit trail, and three outputs: interactive HTML, GraphRAG-ready JSON, and a plain-language GRAPH_REPORT.md.
Use this skill when the user asks to turn code, docs, papers, images, videos, notes, or a mixed corpus into a persistent knowledge graph; query an existing graph; find paths between concepts; add a URL to a graph corpus; export graph artifacts; or run graphify watch/hooks/Claude integration.
Usage
/graphify # full pipeline on current directory -> Obsidian vault
/graphify <path> # full pipeline on specific path
/graphify <path> --mode deep # thorough extraction, richer INFERRED edges
/graphify <path> --update # incremental - re-extract only new/changed files
/graphify <path> --directed # preserve edge direction: source -> target
/graphify <path> --cluster-only # rerun clustering on existing graph
/graphify <path> --no-viz # skip visualization, just report + JSON
/graphify <path> --svg # also export graph.svg
/graphify <path> --graphml # export graph.graphml
/graphify <path> --neo4j # generate graphify-out/cypher.txt
/graphify <path> --mcp # start MCP stdio server for agent access
/graphify <path> --watch # watch folder, auto-rebuild on code changes
/graphify <path> --wiki # build agent-crawlable wiki
/graphify add <url> # fetch URL, save to ./raw, update graph
/graphify query "<question>" # BFS traversal - broad context
/graphify path "AuthModule" "Database" # shortest path between concepts
/graphify explain "SwinTransformer" # plain-language node explanation
Why It Exists
graphify follows the /raw folder workflow: drop anything into a folder — papers, tweets, screenshots, code, notes — and get a structured graph that shows what is connected.
Three things it does that a model alone cannot reliably provide:
- Persistent graph — relationships are stored in
graphify-out/graph.jsonand survive across sessions. - Honest audit trail — every edge is tagged
EXTRACTED,INFERRED, orAMBIGUOUS. - Cross-document surprise — community detection finds relationships across files that the user may not know to ask about.
Use it for new codebases, reading lists, research corpora, and personal raw folders.
Progressive Loading
Read only the file needed for the requested path:
| Request shape | Read next |
|---|---|
Full /graphify pipeline, including install, file detection, and transcription |
SetupAndDetection.md, then Extraction.md, then GraphOutputs.md |
| Entity/relationship extraction details, AST extraction, semantic subagents, caching, or merge logic | Extraction.md |
| Clustering, community labels, HTML, Obsidian, wiki, Neo4j, SVG, GraphML, MCP, benchmark, manifest, and final report | GraphOutputs.md |
--update or --cluster-only |
IncrementalOps.md |
query, path, or explain |
QueryOps.md |
add, --watch, commit hook, or Claude integration |
AutomationOps.md |
| Edge provenance, cost visibility, warning, and reporting constraints | HonestyRules.md |
For normal full-pipeline invocation, follow the referenced files in order and do not skip steps.
Core Rules
- If no path was given, use
.and do not ask for a path. - Never invent an edge. If unsure, use
AMBIGUOUS. - Never skip the corpus-size warning.
- Always show token cost in the report.
- Never hide cohesion scores behind symbols; show the raw number.
- Never run HTML visualization on a graph with more than 5,000 nodes without warning the user.
- Preserve the interpreter guard: subsequent commands must use
$(cat graphify-out/.graphify_python)after setup. - For semantic extraction, use parallel extraction agents as described in
Extraction.md; do not read all files manually one by one.