NetworkX MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
| prompts | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| create_graphC | Create a new graph |
| add_nodesC | Add nodes to a graph |
| add_edgesC | Add edges to a graph |
| get_infoD | Get graph information |
| list_graphsB | List all stored graphs with summary info |
| delete_graphC | Delete a graph from storage |
| remove_nodesC | Remove nodes from a graph |
| remove_edgesC | Remove edges from a graph |
| shortest_pathC | Find shortest path between nodes |
| get_neighborsC | Get all neighbors of a node |
| set_node_attributesC | Set attributes on one or more nodes |
| get_node_attributesC | Get all attributes of a node |
| set_edge_attributesC | Set attributes on one or more edges |
| get_edge_attributesC | Get all attributes of an edge |
| degree_centralityC | Calculate degree centrality for all nodes |
| betweenness_centralityC | Calculate betweenness centrality for all nodes |
| connected_componentsC | Find connected components in the graph |
| pagerankC | Calculate PageRank for all nodes |
| community_detectionC | Detect communities in the graph using Louvain method |
| clustering_coefficientsC | Calculate clustering coefficients for all nodes |
| graph_statisticsB | Calculate comprehensive graph statistics (density, diameter, degree distribution) |
| minimum_spanning_treeC | Find minimum spanning tree of an undirected graph |
| cycles_detectionA | Detect cycles in a graph (cycle basis for undirected, DAG check for directed) |
| graph_coloringC | Color graph vertices using greedy algorithm |
| centrality_measuresB | Calculate multiple centrality measures (degree, betweenness, closeness, eigenvector) |
| matchingC | Find maximum weight matching in a graph |
| maximum_flowC | Calculate maximum flow in a directed graph |
| topological_sortB | Return a topological ordering of a directed acyclic graph |
| subgraphB | Extract an induced subgraph and store it as a new graph |
| merge_graphsB | Compose two graphs into a new graph (union of nodes and edges) |
| visualize_graphC | Create a visualization of the graph |
| import_csvB | Import graph from CSV edge list (format: source,target per line) |
| export_jsonA | Export graph as JSON in node-link format |
| build_citation_networkC | Build citation network from DOIs using CrossRef API |
| analyze_author_impactC | Analyze author impact metrics including h-index |
| find_collaboration_patternsC | Find collaboration patterns in citation network |
| detect_research_trendsC | Detect research trends over time |
| export_bibtexC | Export citation network as BibTeX format |
| recommend_papersC | Recommend papers based on citation network analysis |
| resolve_doiB | Resolve DOI to publication metadata using CrossRef API |
| trigger_workflowC | Trigger a GitHub Actions workflow |
| get_workflow_statusC | Get CI/CD workflow status |
| cancel_workflowC | Cancel a running workflow |
| rerun_failed_jobsC | Rerun failed jobs in a workflow |
| get_dora_metricsB | Get DORA metrics for CI/CD performance |
| analyze_workflow_failuresC | Analyze workflow failures with AI-powered insights |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 46 tools
Multiple tools have overlapping purposes (degree_centrality, betweenness_centrality, and centrality_measures all compute centrality metrics). The inclusion of unrelated domains (citation analysis, CI/CD workflows) alongside graph operations creates significant confusion about tool selection for an agent.
Most tools follow a clear verb_noun snake_case pattern (create_graph, add_edges, delete_graph, trigger_workflow), but a few deviate (pagerank, subgraph, matching, topological_sort). Overall the convention is consistent and readable.
With 46 tools spanning three very different domains (NetworkX graphs, citation networks, CI/CD workflows), the server is excessively large and unfocused. Each domain individually would warrant a smaller, dedicated tool set; combining them makes the count inappropriate.
The graph operations lack obvious essentials such as listing edges or updating graph structure. The citation and CI/CD domains appear arbitrarily bolted on, and while each has some coverage, the overall surface is incomplete for any single coherent purpose.