quickstart

a project that remembers itself

Install it, connect your AI, describe the project. nodestone keeps the plan, the decisions and their reasons, and progress that checks prove. Free for one person, no account.

install

From PyPI. Python ≥ 3.11.

bash

pip install nodestone

nodestone init

Creates the project's records in .nodestone/ inside the project and prints the next steps. Nothing is written anywhere else.

bash

cd your-project
nodestone init

connect your AI

nodestone speaks MCP over stdio. Register it once; Cursor and OpenCode take the same command.

bash

claude mcp add nodestone -- nodestone mcp

Or in an MCP config file:

json

{
  "mcpServers": {
    "nodestone": { "command": "nodestone", "args": ["mcp"] }
  }
}

describe the project

Tell your AI what you are building, in your own words. It plans with nodestone — phases, tasks, dependencies, a definition of done — and asks only what you must decide. Watch it live:

bash

nodestone ui

every day

bash

nodestone today          # your work and why
nodestone start form     # the task is yours
nodestone done form      # nodestone runs the checks and says what is still missing
nodestone report         # the status report, from the records only

a team: Pro

Pro adds people with roles and approvals, and the team server used from other computers. It is priced per person, the owner included.

bash

nodestone activate                      # buy or sign in; the terminal receives the license
nodestone access user-add ayse --role member
nodestone mcp --http --host 0.0.0.0 --port 18766 --tls-cert server.crt --tls-key server.key

# on a teammate's computer
nodestone --server https://team-server:18766 login ayse

pairs with atlas + deepstrain

atlas maps your code and deepstrain changes it — point both at the same project and nodestone keeps the plan and what was proven done. Understand, execute, remember: one loop.

also from us

deepstrain — autonomous execution · atlas — code intelligence

see pricing bundle →