The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external search results and performance tools which provides a surface for instructions that could attempt to influence the agent's behavior.
Ingestion points: External content enters the agent's context through tools like get_search_console_performance and get_serp_results.
Boundary markers: The skill instructions do not specify the use of clear delimiters or instructions to ignore embedded commands in the ingested data.
Capability inventory: The skill can write local files (keywords/ directory), save data to a remote service (save_keywords), and publish content to a public, shareable URL via create_share_report.
Sanitization: There is no mention of sanitizing or escaping content from external web pages or search results before processing or outputting it.
[EXTERNAL_DOWNLOADS]: The skill connects to a remote MCP server at https://app.harborrank.com/mcp to access keyword research functionality.

keyword-research