Show HN: ThoughtDAG – An editable context graph for LLM conversations

ThoughtDAG is a desktop and web application that visualizes LLM conversation context as an editable, directed acyclic graph rather than hiding it in linear chat history. Users can branch conversations, prune irrelevant paths, merge reasoning threads, and inspect exactly what context will be sent to the model before generation, making context selection transparent and controllable. The tool is available for macOS, Windows, and Linux with the same interface across desktop and web versions.