Garphield
Garphield is a free network visualization and analysis workbench. Open CSV, GraphML or GEXF files in your browser with no account, and work by hand or through an agent. Network data leaves the browser only when you send a bad-view report.
A network, in this context, is a set of nodes joined by edges: people and the collaborations between them, papers and their citations, packages and their dependencies, sites and the hyperlinks between them. Garphield lays a network out so its structure is visible, runs standard network-science measures in the browser, binds the results to size, colour and other visual channels, and saves the result as a link, a project file or an image.
Who uses Garphield
Section titled “Who uses Garphield”People with a network to understand and show. Collaboration, citation, social, dependency and infrastructure networks of up to roughly ten thousand nodes. The outcome is a visible structure, a set of identified key nodes, and a figure or a shareable view.
Agents acting for a person. Most workbench operations are also commands
in a published schema. An agent can load, analyse, style and export a network
through llms.txt, a plain-text index of the
docs for language models, and the browser API without using the interface. See
Garphield for agents.
Garphield also fits into existing work:
- Python: display a NetworkX or pandas graph in a notebook and bring edits back. Python
- Embedding: place the renderer in another page with an iframe or the JavaScript driver. Embed Garphield
Start with a question
Section titled “Start with a question”Each link opens the workbench with a dataset loaded and the answer on screen.
- Which characters hold the story together?: Les Misérables, sized by betweenness.
- Which institutions bridge research communities?: EuroSiS web network, coloured by community and sized by betweenness.
- How does a team form over time?: collaboration over time, with a time filter ready to play.
Or open Garphield with your own file, or follow the Quick start.
Ways in
Section titled “Ways in”| Surface | Start here |
|---|---|
| Browser | Open Garphield and drop a file on the welcome dialog. |
| Python | pip install 'garphield[networkx]', then Python. |
| Agent | Point the agent at https://garphield.com/llms.txt, then Garphield for agents. |
| Embed | pnpm add garphield, then Embed Garphield. |
Supported file types
Section titled “Supported file types”Garphield imports common graph and table formats. The smallest useful example of each:
| Format | Extension | Tiny example |
|---|---|---|
| Garphield project | .gph |
A saved Garphield workspace |
| GEXF | .gexf |
<gexf><graph>… |
| GraphML | .graphml |
<graphml><graph>… |
| GML | .gml |
graph [ node [ id 0 ] ] |
| Graphviz DOT | .dot, .gv |
graph { a -- b } |
| Node-link JSON | .json |
{"nodes": […], "edges": […]} |
| CSV or TSV | .csv, .tsv |
source,target |
| Edge list | .edges, .txt |
a b |
| Matrix Market | .mtx |
%%MatrixMarket matrix coordinate… |
See Supported file formats for details.
Contact
Section titled “Contact”Questions, ideas, or something not working? Email [email protected].