Concept detail
two-tier eval suite (R-tasks vs F-tasks)
0 tagged videos · 1 connection videos · degree 0
Publication timeline
- 2026-051 video
Source videos
Observed neighbors
hill-climbing refactoring loop
deterministic regressions guard against breakage while judged failure-mode tasks guide architecture changes
1 source videos
Tool, skill, or subagent? Decomposing an agent that outgrew its promptRelated tagged insights
No core insights are tagged with this concept.