The Agent Skills Directory

[SAFE]: The skill defines a clear instructional workflow for an AI agent to use specific SEO-related tools (MCP servers) to aggregate search performance data.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests external data (search queries and site problems) that could technically contain malicious prompts, it lacks dangerous capabilities such as file system writes, shell execution, or network exfiltration that would allow such an injection to be exploited. The data is processed solely for presentation in markdown tables.
Ingestion points: Data returned from get_yandex_query_stats, get_yandex_site_summary, and get_yandex_indexing_history.
Boundary markers: Not explicitly defined in the prompt instructions.
Capability inventory: Limited to tool calls for data retrieval and generating markdown output.
Sanitization: Not explicitly implemented in the instructions, but the risk is mitigated by the lack of execution capabilities.

yandex-performance