OpenFlowKit 2 has five ways to make a diagram: draw it, write it as code (or paste Mermaid, D2, Structurizr), ask any AI model or a local one, let Claude Code edit it over MCP, or map a GitHub repo with every arrow cited to a file and line.
If you could only keep one, which would it be, and what's missing from it? This decides what I build next.
Every team I've worked on had one: the architecture diagram that was right once, then quietly went stale. OpenFlowKit 2 goes live today, and it's built around keeping that one diagram current: you draw it or write it as code, your AI assistant proposes changes, your coding agent edits it over MCP.
What's yours? The system diagram, the onboarding flow, the database schema? I'd love to try the new version against real cases.
A free, open-source diagram canvas: MIT, local-first, no account, in your browser. One canvas, five ways in. Draw it. Write it as code, or paste Mermaid, D2 or Structurizr. Ask any of 10 AI providers, or a local model. Hand it to your coding agent: Claude Code edits the canvas you have open, live, over MCP. Or pull it from a GitHub repo, every arrow citing its file and line. The CLI checks drift in CI, so diagrams keep up with your code.
OpenFlowKitFree, open-source diagrams you and your AI edit together
