hillclimb demo
Try hillclimb in one command: agents climb the circle-packing problem.
Try hillclimb in one command: agents climb the circle-packing problem.
Creates a hillclimb/ dir here if there is none, installs the bundled problem, starts several searches in parallel in the background, and prints the commands that show them live — run those right here.
hillclimb demo [OPTIONS]Options
| option | what it does |
|---|---|
--budget TEXT | Wall-clock budget per search, e.g. 10m. Default: 10m. |
--parallel-searches INTEGER RANGE | Searches to run at once. Default: 3. |
--parallel-agents, --parallel-operators INTEGER RANGE | Concurrent agents (one candidate each) per search. Default: 3. |
--model TEXT | Model the agent runs, e.g. sonnet / opus. |
--agent, --backend TEXT | The coding agent that runs the operators: claude-code | codex | pi | dummy (--backend is the old spelling). |
hillclimb connect
Connect the agents that run operators (claude, codex, pi) and the OpenRouter route that pays for them.
hillclimb disconnect
Undo `hillclimb connect <target>` on hillclimb's side: unpin it as the default and forget the staged credentials, so the next `connect` sets it up from scratch again.