The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script (keyword_cluster.py) stored within the plugin's root directory. This script performs the core logic for keyword clustering and intent grouping based on provided input files.
[PROMPT_INJECTION]: The skill processes untrusted data from seed keyword CSVs and JSON SERP results, creating a surface for indirect prompt injection. • Ingestion points: Data enters the context from user-provided CSV seeds and external rank-tracker JSON outputs stored in the plugin data directory. • Boundary markers: The skill does not define specific delimiters for the ingested content when drafting pillar page briefs. • Capability inventory: The skill executes a local Python script and writes intermediate and final files to the plugin data directory. • Sanitization: The skill applies a compliance filter to keywords before they are used in the clustering process.

keyword-cluster