The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple external sources which could potentially contain malicious instructions.
Ingestion points: Reads journalctl logs containing raw claims from external Identity Providers (IdP), parses external grafana.ini configuration files, and imports dashboards/overview.json via Terraform.
Boundary markers: None identified; instructions do not include specific delimiters or warnings to ignore embedded content in these data sources.
Capability inventory: The skill has the capability to execute administrative API commands (POST/DELETE) via curl, modify infrastructure via terraform, and read system logs via journalctl.
Sanitization: No explicit sanitization or validation of the content of the JSON dashboards or IdP claims is described before the agent acts on them.
[COMMAND_EXECUTION]: The skill uses standard CLI tools to perform administrative tasks.
Usage of curl for interacting with the Grafana Cloud API is expected and utilizes placeholders for credentials.
Usage of terraform for resource provisioning is a standard practice for this skill's administrative scope.
Usage of journalctl for troubleshooting authentication flows is appropriate for the 'admin' persona.
[SAFE]: The skill demonstrates safe handling of sensitive information.
Hardcoded secret placeholders: Uses <grafana-com-api-key>, your_client_secret, and var.grafana_service_account_token instead of real credentials.
Trusted Sources: All external references target official Grafana domains (grafana.com, grafana.net) and the official Grafana Terraform provider, which are verified vendor resources.

admin