The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (suggest.py, firecrawl.py, pageviews.py) located within the ${CLAUDE_PLUGIN_ROOT}/scripts/connectors/ directory to perform data harvesting and analysis tasks.
[EXTERNAL_DOWNLOADS]: The skill interacts with external services to gather research data, including Google's autocomplete endpoint, Wikipedia's pageview statistics, and the Firecrawl search API. These references to well-known technology services are standard for SEO research and do not represent a security risk.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes and interprets data from external sources (search results and autocomplete suggestions).
Ingestion points: Data retrieved from external SEO connectors during the 'Discover' and 'GEO-Check' phases.
Boundary markers: No explicit delimiters are specified to separate external content from system instructions in the prompt templates.
Capability inventory: The skill's primary capabilities include executing shell-based data connectors and writing research deliverables to the local filesystem (memory/research/).
Sanitization: No specific sanitization or filtering logic is mentioned for the external data before it is presented to the agent.

keyword-research