D2 vs tldraw SDK + MCP

D2

A single local command turns text into a diagram in six formats with no account, key or network call, and CI passes on the default branch. The language is at 0.9.0 with no stable release, its company sponsor has shut down, and one maintainer with limited time wrote every commit since July.

tldraw SDK + MCP

Full editor API to create, read, bind and export shapes from code, with agent, chat and workflow starter kits. Source-available, and commercial prices aren't published.

D2

1. Check the exit status of d2, never the output file. The man page says a partial render can be written when an error occurs
2. Run d2 validate file.d2 before rendering, and d2 fmt to normalise the source
3. Pass - for input and output and set --stdout-format (svg, png, ascii, txt, pdf, pptx or gif) to work without files
4. Use --stdout-format ascii when the reader is a text-only model
5. Pass --bundle=false or review image URLs and imports before rendering D2 text from an untrusted source. Release 0.9.0 fetches remote images when bundling

tldraw SDK + MCP

1. Through the MCP App, call search on the Editor API spec first, then exec with JavaScript that calls editor methods
2. Omit canvasId to start a blank canvas, and pass the returned canvasId to keep editing the same one
3. Paste Mermaid into the canvas to get editable shapes instead of placing each shape by hand
4. Return editor.getCurrentPageShapes() from exec to check what was drawn
5. Read the release notes before a minor upgrade. Minor versions can break