Consensus-AI turns a draft into a plan your LLM can build, one settled question at a time.
Working a design out with an LLM is fast, until it is not.
You settled that question an hour ago. Now it is forty messages up, and the model has forgotten it.
One sentence read two ways. The model picked one, built it beautifully, and now you are unpicking it.
The plan that was built is not the plan you have. The reasons are somewhere in a conversation nobody will reread.
A plan is a document with threads. Each thread is one question, settled in one place, and applied to the text in front of you.
Markdown, in an editor with a live preview. Rough is fine: the draft is where the questions come from, not the answers.
Select a passage and comment. The LLM replies. You resolve. It applies the outcome to the text, and the thread turns green with the revision it made.
The plan is done when every thread is applied. Say "build it": your agent implements exactly that plan, and its build log is a thread too, accepted the same way.
A record you can trust, and an LLM that only does what you settled.
On your machine, on a server of yours, or in a window of its own.
One project folder in one window, with its own server inside for you and your agent. No account needed; signed in, it connects to Servers.
One server for several people, in a window of its own or headless: the projects, the users and their roles, the LLM connections and tasks.
One image, one volume, port 3500. Put it on a box on your network and work from anywhere on it.
The code is public at github.com/mchiver/consensus-ai. Four ways to run it.
Clone, install, and start the Server.
git clone https://github.com/mchiver/consensus-ai.git
cd consensus-ai
npm install
npm run server
Sign in on its page with a Personal license to become its Administrator. More in the guide.
Build the image from the checkout; the data lives in its volume.
docker build -t consensus:latest .
docker compose up -d
Port 3500. More in the guide.
Run the desktop from the checkout and open a folder, or connect to a Server.
npm run desktop
Consensus Desktop and Consensus Server, each packaged with its server: download, install, open. The installers are not signed yet, so Windows asks once before each runs.
Consensus-AI is not Consensus, the research search engine at consensus.app, and has no connection to it. Consensus-AI is a tool for settling a plan with your LLM, thread by thread.