CharDesk
Render and edit diagrams, interfaces, and visual slides directly on an infinite Unicode character grid shared between humans and AI agents.
CharDesk provides a shared visual workspace for humans and AI agents powered by visual Unicode text. Combining an infinite canvas, box-drawing characters, Nerd Fonts, and ANSI styling, it represents spatial layouts, system architectures, data plots, and interfaces as clean, structured text. Because scenes remain text-based rather than rasterized images, LLM agents can inspect, search, diff, and update layouts across turns with pinpoint accuracy. The integrated Model Context Protocol (MCP) server allows AI assistants like Claude Code and Codex to automatically launch the canvas, generate visual explanations, and co-author character-based presentations and designs in real time.