The Agent Skills Directory

[SAFE]: The skill's primary function is to perform comparative analysis of search engine performance data. All operations are conducted through a defined set of MCP tools (list_sites, get_ranking_changes, etc.), ensuring access is limited to the intended data scope.
[SAFE]: No evidence of prompt injection or bypass attempts was found. The instructions are clearly defined and focused on data interpretation and reporting.
[SAFE]: There are no remote code execution patterns, external package dependencies, or obfuscated contents within the skill.
[SAFE]: The skill includes strong grounding requirements, explicitly instructing the agent to never report numbers that weren't returned by a tool and to site the source tool for all metrics.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data from Google Search Console (metrics and annotations). While this is technically an external ingestion point, the risk is negligible as the agent uses this data solely for generating reports and tables, and the skill lacks any capabilities (like shell access or file writing) that could be exploited through malicious data content.

ranking-changes