Get Confidence
Get confidence metrics for an experiment.
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Path Parameters
Response
Successful Response
Confidence metrics for an experiment.
overall_confidence is the headline DECISION confidence (how sure we are
the recommended configuration beats a typical one). signal_strength is
the PPC predictability sub-metric (null-safe honesty guard). The per-claim
parameter_report + recommended_config carry the "big picture".
Sub-metric breakdown (decision/signal/skill/depth/breadth)
Headline decision confidence (0-100)
0 <= x <= 100Axis metadata (axis_key/label/description/weight)
Weighted combined recommended config across axes
90% CI for decision confidence
True iff the experiment defines explicit voting axes
Per-parameter effects/curves with per-claim confidence chips
Synthesized recommended configuration
Per-axis confidence report keyed by axis_key
PPC predictive skill 0-100 (null-safe guard)

