{"slug": "show-hn-thoughtdag-an-editable-context-graph-for-llm-conversations", "title": "Show HN: ThoughtDAG – An editable context graph for LLM conversations", "summary": "ThoughtDAG, an editable context graph for LLM conversations, launched as a Show HN project with a web demo and desktop app, allowing users to branch, edit, and inspect the context sent to language models. The desktop app includes keyless web search and local storage, with version v0.3.13 available on GitHub Releases.", "body_md": "**Branch**\n\nExplore another interpretation without overwriting the path that led you here.\n\nThoughtDAG turns every question-and-answer exchange into a node, and every connection into context. Remove one edge, ask the same question again, and that branch is gone from the request.\n\nThe web demo is a feature subset for a quick look — the example canvas needs no key. The desktop app is the full instrument: keyless web search, every connection tool, and your canvases living on your own machine.\n\nThoughtDAG walks the incoming edges of a node, orders the relevant ancestors, and builds the message sequence sent to the selected model. The structure stays visible, editable, and inspectable.\n\nExplore another interpretation without overwriting the path that led you here.\n\nKeep a useful detour on the canvas while excluding it from the next request.\n\nBring selected evidence and reasoning paths back together in one answer.\n\nPreview what the model will receive before generation. No hidden memory selection.\n\nThe same app in its own window, local engine bundled. No Node, no terminal.\n\nv0.3.13\nAll versions and history live on [GitHub Releases](https://github.com/chenxiachan/thoughtdag/releases). The desktop and web apps are the same app; your canvases stay on your device either way.\n\nThe history is here. Which parts enter the next request?\n\nThis history should not enter the research summary.\n\nStill connectedThe prompt stayed the same. Polluted context changed the answer.\n\nIncludes unrelated branchIncoming ancestors:\n\nResearch question Evidence A Dinner detourAfter deleting the orange edge: −47 tokens\n\nThe words are identical. Only one edge changed.\n\nReproducible contextThe unrelated dinner suggestion is gone.\n\nAnswer updated in placeNo hidden memory selector. What the model sees, why, and what was removed stay visible in the graph.\n\nVisible Editable Inspectable", "url": "https://wpnews.pro/news/show-hn-thoughtdag-an-editable-context-graph-for-llm-conversations", "canonical_source": "https://chenxiachan.github.io/thoughtdag/", "published_at": "2026-08-15 04:42:07+00:00", "updated_at": "2026-08-15 05:10:46.667788+00:00", "lang": "en", "topics": ["ai-tools", "large-language-models", "developer-tools"], "entities": ["ThoughtDAG", "GitHub Releases", "chenxiachan"], "alternates": {"html": "https://wpnews.pro/news/show-hn-thoughtdag-an-editable-context-graph-for-llm-conversations", "markdown": "https://wpnews.pro/news/show-hn-thoughtdag-an-editable-context-graph-for-llm-conversations.md", "text": "https://wpnews.pro/news/show-hn-thoughtdag-an-editable-context-graph-for-llm-conversations.txt", "jsonld": "https://wpnews.pro/news/show-hn-thoughtdag-an-editable-context-graph-for-llm-conversations.jsonld"}}