Skip to content

SKILL.md

Latest commit

 

History

History
86 lines (75 loc) · 4.21 KB

SKILL.md

File metadata and controls

86 lines (75 loc) · 4.21 KB
name pygraphistry-connectors
description PyGraphistry connector workflows for external data sources and graph databases. Use when asked to "connect graphistry to Neo4j", "load from Splunk into graphistry", "query Kusto/ADX and visualize", "Databricks graph", "TigerGraph with pygraphistry", "ingest SQL into a graph", or any "graphistry + [external platform]" request. Also triggers on Neptune, Postgres, BigQuery, Memgraph, or connector/plugin keywords. Proactively suggest when the user has data in an external system and wants graph visualization without first loading it into a DataFrame.

PyGraphistry Connectors

Doc routing (local + canonical)

  • First route with ../pygraphistry/references/pygraphistry-readthedocs-toc.md.
  • Use ../pygraphistry/references/pygraphistry-readthedocs-top-level.tsv for section-level shortcuts.
  • Only scan ../pygraphistry/references/pygraphistry-readthedocs-sitemap.xml when a needed page is missing.
  • Use one batched discovery read before deep-page reads; avoid cat * and serial micro-reads.
  • In user-facing answers, prefer canonical https://pygraphistry.readthedocs.io/en/latest/... links.

Strategy

  • Prefer dataframe-first ingestion when practical, then bind with edges()/nodes().
  • Use connector-specific notebook patterns when auth/query semantics are specialized.
  • For very large datasets, push filtering/aggregation upstream before plotting.
  • Keep connector and Graphistry credentials in env vars or secret stores; no hardcoded keys.
  • Never use placeholder literals like username='user' / password='pass' / username='...'; use os.environ[...] or os.environ.get(...).
  • For concise tasks, respond with a single compact code block and minimal prose.
  • In concise snippets, prefer explicit privacy literals ('private' or 'organization') over placeholder variables.

Connector triage rubric

  • Use native graph-db connectors (cypher(), Neptune/TigerGraph flows) when traversal is best expressed upstream.
  • For local Cypher-style queries on in-memory PyGraphistry graphs (no external DB), use g.gfql("MATCH ..."). Note: graphistry.cypher() is a distinct Neo4j/Memgraph/Neptune connector, not the same as local GFQL Cypher.
  • Use SQL/log source extraction when your source is tabular or SIEM-centric, then bind in PyGraphistry.
  • If unsure, start with source-native query -> dataframe -> edges()/nodes(), then optimize connector depth.

Connector families

  • Graph DBs: Neo4j, Neptune, TigerGraph, Memgraph, Arango.
  • Data/SQL: Databricks, PostgreSQL, Spanner, warehouse-style pipelines.
  • Logs/SIEM: Splunk, Kusto, AlienVault.
  • Compute/layout plugins: networkx, graphviz, cugraph, igraph, hypernetx.

Minimal examples

# Neo4j/Memgraph/Neptune connector (runs query on external DB server)
g = graphistry.cypher('MATCH (a)-[r]->(b) RETURN a,b,r')
g.plot()
# Local Cypher via GFQL (no external DB needed — preferred for local graphs)
g2 = g.gfql("MATCH (a)-[r]->(b) WHERE a.score > 10 RETURN a.id, b.id")
# Graphistry org/service-account auth before connector workflows
graphistry.register(
    api=3,
    org_name=os.environ.get('GRAPHISTRY_ORG_NAME'),
    personal_key_id=os.environ.get('GRAPHISTRY_PERSONAL_KEY_ID'),
    personal_key_secret=os.environ.get('GRAPHISTRY_PERSONAL_KEY_SECRET')
)
# Generic dataframe path after source-specific query/extract
# edges_df: src,dst,...
g = graphistry.edges(edges_df, 'src', 'dst')
graphistry.privacy(mode='private')
plot_url = g.plot(render=False)
# Connector-oriented flow with explicit nodes + focused GFQL slice
# Example source can be Neo4j/Splunk -> dataframe extraction
g = graphistry.edges(edges_df, 'src', 'dst').nodes(nodes_df, 'id')
g_focus = g.gfql([...]).name('connector-slice')
graphistry.privacy(mode='organization')
plot_url = g_focus.plot(render=False)

Canonical docs