π¦ ClawHub
Skill Graphify
by @flobo3
Turn any folder of code, docs, papers, or images into a queryable knowledge graph. Cross-platform wrapper for graphify CLI.
β‘ When to Use
π‘ Examples
Step 1 β Ensure graphify is installed
python graphify_wrapper.py ensure-installed
Or manually: pip install graphifyy
Step 2 β Build knowledge graph
python graphify_wrapper.py build /path/to/project
This runs the full pipeline: detect files β AST extraction β build graph β cluster β export.
Output goes to
graph.html β interactive visualization (open in browser)GRAPH_REPORT.md β plain-language audit reportgraph.json β queryable knowledge graphcache/ β SHA256 cache for incremental updatesStep 3 β Read the report
python graphify_wrapper.py report
Or read graphify-out/GRAPH_REPORT.md directly. Present the key findings to the user: god nodes (highly connected), surprising connections, community structure.
Step 4 β Query the graph (optional)
python graphify_wrapper.py query "how does authentication work"
Or use the CLI directly for more options:
graphify query "show the auth flow" --graph graphify-out/graph.json
graphify query "what connects X to Y?" --graph graphify-out/graph.json --dfs
graphify query "explain dependency injection" --budget 1500 --graph graphify-out/graph.json
Send results to user
After building, send graphify-out/graph.html to the user so they can explore the interactive graph. Summarize GRAPH_REPORT.md in your response.
π Tips & Best Practices
graphify_wrapper.py) handles cross-platform compatibility (Windows CMD, Linux, macOS).graphifyignore file (same syntax as .gitignore) to exclude directoriesTERMINAL
clawhub install skill-graphify