generated: '2026-07-19'
method: searched
status: published
official: false
source: https://github.com/graphistry/graphistry-mcp
server:
  name: graphistry-mcp
  maintainer: community (silkspace)
  official: false
  transport:
  - stdio
  - http
  install:
  - npx -y @silkspace/graphistry-mcp
  - python run_graphistry_mcp.py
  - ./start-graphistry-mcp.sh --http 8080
  requires: A free Graphistry Hub account; credentials supplied via environment variables or a .env file.
  description: GPU-accelerated graph visualization and analytics for LLMs using Graphistry and MCP. Community-maintained
    (not a first-party Graphistry product).
tools:
- name: visualize_graph
  description: Render a graph or hypergraph in Graphistry.
- name: get_graph_ids
  description: List stored graph IDs.
- name: get_graph_info
  description: Retrieve graph metadata.
- name: apply_layout
  description: Apply a layout algorithm to a graph.
- name: detect_patterns
  description: Run network/pattern analysis.
- name: encode_point_color
  description: Encode point color.
- name: encode_point_size
  description: Encode point size.
- name: encode_point_icon
  description: Encode point icon.
- name: encode_point_badge
  description: Encode point badge.
- name: apply_ring_categorical_layout
  description: Apply a categorical ring layout.
- name: apply_group_in_a_box_layout
  description: Apply a group-in-a-box layout.
- name: apply_modularity_weighted_layout
  description: Apply a modularity-weighted layout.
- name: apply_ring_continuous_layout
  description: Apply a continuous ring layout.
- name: apply_time_ring_layout
  description: Apply a time-ring layout.
- name: apply_tree_layout
  description: Apply a tree layout.
- name: set_graph_settings
  description: Configure visualization parameters.
deployment:
  mode: local-stdio
  verified: searched
  tools: 16
  checked: '2026-08-12'
  source: catalog MCP census