The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from a LeanIX workspace, which constitutes an untrusted external ingestion surface. Ingestion points include the mcp__leanix__filter_inventory and mcp__leanix__search_inventory tools. The skill instructs the agent to perform structural verification of data candidates and prioritize structured ID lookups over semantic search to mitigate risks. It utilizes mcp__leanix__execute_code for processing this data.
[DYNAMIC_EXECUTION]: The skill uses mcp__leanix__execute_code to perform data analysis tasks that require logic beyond simple queries. Reference files contain JavaScript templates for the agent to adapt and run in the code sandbox for tasks such as calculating completion percentages, ranking results, and traversing complex relations.
[COMMAND_EXECUTION]: The skill leverages the mcp__leanix__execute_code tool as a primary mechanism for data orchestration and multi-step inventory workflows.

fact-sheet-search