feature · learning-adaptation
Taxonomy
Extrapolated — agenticness
5 tags · 38 out · 38 in · 3 artifacts · 1 gaps · 1 corpus docs
Learning & adaptation is the ability of an agent system to improve or modify its behavior from feedback, evaluations, or experience — beyond a human editing a static prompt pack. The Agent Compendium treats this as one of five dimensions of agenticness: performance improvement and behavior modification over time.
Forms include offline fine-tuning hooks, online preference updates, self-reflection loops that change durable policies, and eval→deploy pipelines that automatically promote better variants. In production, “learning” without promotion control is indistinguishable from silent drift.
Canon forbids silent company-shape change. Learning that alters thresholds, routing, or authority must be a reviewed diff — or it is a product bug. Measured adaptation under gate is valuable; unsupervised policy mutation is not.
Graph edges for this feature — technologies, category lens, companies, DM.
features/learning-adaptation/README.md
feature missing document-management.md
Learning & Adaptation · ok
features/learning-adaptation/README.md
Research path: features/learning-adaptation