Use case
Architecture decision
Propose an approach and API contract with the trade-offs and risks made explicit — for a human to approve before anything is built.
What it is
The stage, and what it produces
A technical agent proposes architecture, trade-offs and the API contract, grounded in your existing decisions and constraints. The value isn't a confident answer — it's a decision a human can weigh, approve and have remembered.
In a run
How it works, step by step
Propose an approach, with the trade-offs stated
Write the API contract — endpoints, payloads, errors
Flag technical risks and constraints
A human approves; the decision and why are kept
Illustrative — produced, reviewed and approved before the next step.
Works with
The parts of the platform it leans on
Where it stands. Part of the Planned orchestration. The Memory that keeps decisions and their reasoning ships today.
Turn this stage into a governed output
Start with what ships today, and grow into the full run as it lands.