author-learnings

name	author-learnings
description	Read a delivered unit's real outcomes — the measure lens (baseline/target/realized), the validate verdicts and fix reports, the run lens (production actuals), and the delivered epic/slice status — against the current product model, and author the MEANING updates the delivery taught — a refined capability/functionality one_line, a raised nfr_needs level (monotonic-up), an earned status promotion, the grounding-doc sections the learning changed, and a new append-only decision per material learning. Each proposed change carries the outcome that justifies it and a confidence tier. Writes the rewritten grounding docs STRAIGHT to the live model (per-node docs only) and emits the spine meaning-field deltas and the decision records as structured data in an STM learn-manifest.yaml — never the shared model files (_spine.yaml, profile, decisions), never the tree skeleton, never a slice/epic entry. Generative artifact production for the /learn play.
version	0.2.0
user-invocable	false
model	opus
allowed-tools	Read, Write, Bash, Glob

author-learnings

Turns a into the updates that keep the living product model honest. After a unit ships, the pipeline has already produced the evidence — the measure lens (each metric's baseline → target → realized + proof), the validate verdicts and fix reports (which gates truly cleared; a is a model-was-wrong signal), the run lens (production actuals vs planned), and the delivered epic/slice status. This skill reads that evidence against what the model , and authors the precise updates the delivery taught — never the tree skeleton, never a slice or epic entry. It is the close of the loop the old enrich gate used to be.