The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches plugin metadata, configuration schemas, and examples from Grafana's official domains (grafana.com and grafana.net). These are trusted vendor resources used to ensure accurate configuration generation.
[DATA_EXFILTRATION]: The skill follows security best practices by instructing users to use environment variable placeholders (${ENV_VAR}) or sensitive Terraform variables for credentials, preventing the exposure of secrets in configuration files or chat history.
[COMMAND_EXECUTION]: Provides standard command-line examples using curl and jq to interact with Grafana APIs. These commands are transparent and target official endpoints.
[PROMPT_INJECTION]: The skill does not contain instructions that attempt to override agent safety guidelines or extract system prompts.
[DATA_EXPOSURE]: Includes explicit warnings against asking users to provide sensitive authentication tokens directly in the chat, relying instead on secure local execution or managed MCP servers.

datasource-provisioning