Sample
ESP32-C3 Example
The fastest way to see ContextGraph::ai end to end is to start from a finished map. This sample is a component-driven map of an ESP32-C3 microcontroller driving a 16-pixel WS2812B strip with a scrolling rainbow — every GPIO pin is captured as a function, and the canvas AI Context holds the full PlatformIO build spec.
1
Download the sample
Open the Downloads page and grab
ESP32-C3-Example.dgrm.md. It is a normal saved map — Markdown with the diagram state embedded.
2
Open it in the tool
In the tool, click ↑ Load, choose the file, and pick Open as separate. The map appears on the canvas.
3
Explore the components
This is a component-driven map: groups → components → functions. The ESP32-C3 component lists every chip/board pin as a function (with strapping- and flash-pin caveats); WS2812B Strip is the LED hardware; Firmware holds the FastLED driver, the rainbow effect, and setup/loop.
4
Read the AI Context
Open the canvas AI Context (and the node notes). It carries the binding build spec: board, wiring (GPIO8 → 330Ω → DIN, common GND, external 5V), the full
platformio.ini, and a src/main.cpp outline for the rainbow.
5
Build it with your agent
Export ▾ → ⇡ Share (or ↗ Export for AI Build) and hand it to your agent — it has everything it needs to generate the PlatformIO project and flash it.
Try asking your agent
▸ The MCP examples assume the contextgraph server is connected — see Setting up your MCP. The paste-a-link examples need no setup.
With the MCP server connected
claude "Read my ESP32-C3 WS2812B Rainbow diagram and generate the full PlatformIO project — platformio.ini and src/main.cpp"Or paste a public link — no MCP needed
claude "From https://ctxgraphai.com/v/ab12cd34.md, scaffold the PlatformIO project and implement the scrolling rainbow on GPIO8"