The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The documentation provides instructions for installing the networkx library and its optional dependencies using the uv pip package manager. These installation steps target the official Python Package Index (PyPI), which is a trusted repository.
[REMOTE_CODE_EXECUTION]: The references/io.md file describes data persistence methods using the pickle module (pickle.load and nx.read_gpickle). Deserialization using pickle is inherently insecure as it can lead to arbitrary code execution if the graph files are sourced from untrusted parties.
[PROMPT_INJECTION]: The skill facilitates the ingestion of data from various external formats (GraphML, GML, JSON, CSV, SQL), which constitutes an attack surface for indirect prompt injection.
Ingestion points: Data is read from external files using functions like nx.read_graphml, nx.read_gml, and pd.read_sql_query as documented in references/io.md.
Boundary markers: The skill does not define boundary markers or include instructions for the agent to ignore potentially malicious embedded content within the graph data.
Capability inventory: The agent has capabilities to write files (nx.write_*), export data to CSV, and generate interactive HTML visualizations (net.show).
Sanitization: There are no suggested validation or sanitization routines for the data before it is processed by the graph algorithms.

networkx