MCP server · Notes
ThoughtDAG MCP server
by chenxiachan
Let your AI search your past chat history by file name or phrase and pull the exact old turns back into the conversation.
ThoughtDAG is a tool that keeps a searchable record of your past AI conversations, and this MCP server is the part that lets your AI search that record for you. Instead of scrolling through old chats trying to remember where you discussed a file or a phrase, you just ask, and your AI looks it up. It is handy for anyone who has been using AI assistants for a while and keeps losing track of what was said when.
What is an MCP server? The 30-second version
On its own, your AI can only chat with you about what is in front of it right now. An MCP server is a small helper program that gives your AI one extra skill, in this case the ability to search your local ThoughtDAG history. Once it is connected, your AI can look things up in that history when you ask, instead of you having to find them yourself. It is read-only, so it only looks things up, it does not change or delete anything.
What this MCP server does
You ask your AI something like which conversations mentioned a certain file, or where a phrase came up. Your AI passes that question to this helper program. The helper searches the local ThoughtDAG index of your past agent sessions and canvases. It finds the matching turns and hands them back to your AI. Your AI then shows you the results in the chat, with the original source turns so you can see where they came from.
What you can do with it
- Find past conversations that mention a specific file
- Search your history for an exact phrase you remember
- See the topics that show up in your local index
- Pull the original turns from an old session back into your current chat
- Jump to a matching turn without opening the desktop app
- Check which sources a recalled excerpt came from
Try asking your AI
- “Which of my past conversations talked about src/lib/api.ts?”
- “Find the conversation where I discussed the phrase retry on timeout.”
- “What topics are in my local ThoughtDAG index?”
- “Show me the original turns from the session where I planned the login flow.”
What it gives back to you
It gives back a list of matching conversation turns, each with the text and a link back to the source session or canvas. You also get topic names from your local index when you ask for those. The results appear as a short list in the chat, so you can read them and decide what to bring into your next question.
Before you start
What you need
- Node.js installed on your computer
- ThoughtDAG installed and set up with some conversation history to search
- Run thoughtdag setup mcp once to connect the history tools to your AI
Good to know
It reads your local conversation history, so anything you search for is visible to your AI, and the README notes that retrieval can miss relevant history and that generated dossiers still need checking.
Install it with your AI
Add ThoughtDAG MCP server to your AI, no technical skills needed
You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.
Sign in to get the install prompt
Members get a ready-made prompt that lets the Claude desktop app check ThoughtDAG MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,611 MCP servers here for you.
Who it's for
People who use AI assistants often and want to find and reuse what they discussed in earlier sessions, especially developers and anyone working on long projects.
