skip to content
~/docs~/integrations~/skills~/faq

</aethereum> / for / solo

Solo, many agents

One developer, many agents, one honest contract

Running Claude Code on the backend and Cursor on the frontend is a fast solo setup, until the two halves stop agreeing on the API between them. Aethereum makes that one contract explicit and shared, so when the backend agent reshapes an endpoint, the frontend agent is told before it writes against the old shape. You stay the human in the loop, but you are no longer the only memory of the interface.

where it breaks

The seam moves

The backend agent renames an endpoint or changes a field; the frontend agent keeps calling the old shape.

You are the only memory

The two agents do not share a context window, so only you remember the exact shape of every endpoint as it changes.

It surfaces at runtime

Each side is internally correct, so the mismatch only shows up where they meet, usually after you have moved on.

how it works
  1. Run one command in the project. It wires Claude Code and Cursor to a free room, no signup.
  2. The backend agent declares each endpoint's shape as it builds it.
  3. The frontend agent reads that exact shape and is alerted the moment it changes.
  4. Keep using both agents exactly as you do now.
npx aethereum init

On a single machine the win is the shared contract between your backend and frontend agents, which do not otherwise see each other. Across a laptop and a desktop, the same room keeps them in sync too. Only the contract and a one-line intent are shared, never your source.

get started free →read the solo guidefor teams