I built Codag because I kept getting lost in my own AI code.

You're chaining 3 LLM calls across 5 files. A prompt change breaks something downstream. Which call? Which branch? You grep for "openai.chat", open 8 tabs, trace the flow manually.

Codag automates this: - Point it at your codebase and it extracts every LLM call, decision branch, and processing step - Renders an interactive and shareable DAG with clickable nodes that link back to source - Live updates as you edit using tree-sitter — no waiting for re-analysis

Supports OpenAI, Anthropic, Gemini, LangChain, LangGraph, CrewAI, and more. Works with Python, TypeScript, Go, Rust, Java, and more.

It's a VS Code extension + self-hosted backend (Gemini 2.5 Flash for analysis). Open source, MIT licensed.