cluster-auto-k-heuristic
IN premise
Auto cluster count is computed as `len(beliefs) // 5`, clamped between 2 and `min(budget // 3, 20)`, targeting approximately 5 beliefs per cluster with at least 3 beliefs per cluster given the budget.
Summary
The system automatically decides how many topic clusters to create by aiming for roughly 5 beliefs in each cluster. It scales with the number of beliefs but stays within guardrails — never fewer than 2 clusters and never more than 20 or a third of the budget, whichever is smaller. This means cluster granularity adjusts naturally as the knowledge base grows, without manual tuning.
Details
| Source | entries/2026/05/08/reasons_lib-cluster.md |