Draft. Work in progress: an outline with working notes. The finished post will replace it.
Dynamic delegation: letting agent topology follow the problem
Fixed multi-agent org charts may be the wrong abstraction. What if structure formed around each task instead?
Most multi-agent systems start with an org chart: a planner, a coder, a reviewer, a tester. The structure is decided before the problem is known. SOAR starts from the opposite end: give an agent a goal and a plan, and let it decide whether the work fits in one unit or needs to be split.
The core loop
- Receive goal + plan.
- Can this be completed as one unit of work?
- If yes, do it. If no, decompose, and delegate each part to a worker built for it.
- Workers apply the same loop, recursively.
[TODO: diagram of a task tree forming and collapsing]
Why fixed hierarchies struggle
- Some tasks are one step; some are hundreds. A fixed shape fits neither well.
- [TODO: an example task where a fixed pipeline wasted effort or missed a dependency]
The hard parts
- Context routing: what does each worker need to know, and nothing more?
- Task-local memory: state that belongs to a subtask, not the whole run.
- Healing: what happens when a worker fails or wanders off.
- Verification: "done" has to be checked, not declared.
- [TODO: which of these is solved, which is still open]
Early observations
[TODO: what has worked and what hasn't. No numbers until they're measured properly.]
Where it goes next
[TODO: link to the spec / repo if it goes public]