The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the elastic CLI to interact with Elasticsearch and Kibana clusters. This is the intended and primary purpose of the skill. The instructions map standard HTTP API calls to specific CLI commands to ensure consistency and security.
[DATA_EXPOSURE]: The skill reads cluster information such as license status, machine learning capacity, and package metadata. It specifically instructs the agent not to guess credentials and to verify the user has the necessary permissions (e.g., manage_ml) before proceeding.
[INDIRECT_PROMPT_INJECTION]: The skill includes functionality to read README files from the Fleet package registry and search results from Elasticsearch indices. While this represents a data ingestion surface, the skill implements strict controls, such as mapping user requests to a static capability table in references/package-capabilities.md and requiring explicit user confirmation before any state-changing actions are taken.
[EXTERNAL_DOWNLOADS]: The skill references the official Elastic CLI repository on GitHub for configuration and prerequisites. These references target a trusted vendor and are documented neutrally as required setup steps.

security-ml-integrations