build-dashboard
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- Data Ingestion Surface: The skill is designed to gather business context and data from multiple sources, including structured databases (~~structured_data), company documents (~~company_docs), and team communications (~~team_communication). This enables the creation of context-aware dashboards but requires the agent to process data from external, potentially untrusted sources to define metrics and requirements.
- Dynamic Dashboard Execution: To support interactive dashboards, the skill generates and executes code for platforms like Streamlit and static HTML. This involves creating Python scripts or HTML/JavaScript files. The workflow includes a validation step where the agent is instructed to launch and test these applications locally (e.g., using
streamlit run) to ensure they render correctly before handoff. - System Artifact Interaction: The skill utilizes a first-party MCP artifact application to render dashboards. This process involves generating structured JSON manifests and datasets. These artifacts are validated using specialized tools (
validate_artifact) to ensure they meet the platform's safety and rendering schemas.
Audit Metadata