The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill's content is limited to technical documentation, LookML configuration snippets, and dashboard templates.
[PROMPT_INJECTION]: The skill provides instructions for using HTML tags such as <font>, <a>, <nav>, and <svg>, as well as images from well-known external sources in dashboard tiles. While these are standard Looker features for branding and navigation, they represent a potential surface for indirect prompt injection or Cross-Site Scripting (XSS) if the dashboard elements are configured to render untrusted dimension data from the underlying database.
Ingestion points: LookML parameters body_text, title_text, and subtitle_text in SKILL.md, references/boilerplate_dashboard.md, and references/markdown_html_templates.md.
Boundary markers: Absent; the templates do not include specific delimiters or instructions for handling dynamic data interpolation.
Capability inventory: Looker dashboard rendering engine (supports a subset of HTML and Markdown).
Sanitization: Absent; the design guidelines do not provide instructions for sanitizing or escaping data values before they are rendered in HTML-enabled tiles.

lookml-dashboards