Microsoft has developed Flint, a visualization intermediate language designed to enable AI agents to transform high-level, human-readable semantic specifications into sophisticated, polished visualizations. Instead of requiring agents to manually configure verbose parameters for axes, scales, and layouts, Flint utilizes a semantic specification layer featuring over 70 data types, such as "Price," "Temperature," and "Country," to automate design decisions. The system employs a dual-layer approach where a compact chart specification defines the data meaning, while a formal ThemeSpec defines the visual identity, allowing for consistent application of styles like "Economist," "Swiss," or "McKinsey" across different platforms. The architecture supports multiple backends, compiling a single input into native specifications for Vega-Lite, ECharts, Chart.js, Plotly, or editable Excel charts. To facilitate agentic workflows, the project includes flint-chart-mcp, a Model Context Protocol server that allows LLMs in environments like GitHub Copilot or Claude to create, validate, and render interactive charts directly. Recent updates have expanded the library's capabilities, including the addition of 38 Plotly chart types, 18 native Excel templates, and expanded theme support for Plotly as of version 0.5.1.